Patent Application
20250026579
The present invention relates to a conveyor controlling apparatus and a conveyor controlling method for controlling a conveyor in accordance with environment parameters with respect to the conveyor.
In recent years, automatic systematisation in logistics has been steadily progressing, and technologies related to automating warehouse management and handling of conveyance items such as checked-luggage and parcels at airports have been developed. The technologies related to automated systems in logistics include, for example, U.S. Pat. No. 5,165,520 Publication, US Patent Application Publication No. 2018/0148271 and U.S. Pat. No. 7,055,672 Publication.
Nevertheless, it is still common to input conveyance items to automatic conveyance systems by hand. For example, in conventional airports, it is common that the checking-in of passengers' luggage is manually handled by ground handling staff. However, even if the ground-handling staff are well trained, there will invariably by situations where the luggage is presented in an inappropriate configuration, or portions of the luggage are in an awkward position (for example, the handle of a suitcase may be extended, or the zipper on a duffel bag may open). This problem is also prevalent in self-drop luggage kiosks.
The checked-luggage can be in transit on a baggage handling system (BHS) and on its way to being loaded onto a departing plane. Therefore, to ensure that the loading of the checked-luggage is seamless and does not cause any delay to the departing plane, it is vital to ensure that further checks and measures are in place to ensure that all checked-luggage satisfy a ‘forwarding condition’ before being loaded onto the plane.
The forwarding condition is mainly subject to capability of the BHS. Accordingly, if a conveyance item does not meet the forwarding condition, it may cause problems in the BHS such as conveyance jams.
Thus, a solution is required that automatically detects any checked-luggage that does not satisfy the forwarding condition. This solution must be capable of clearly identifying the problematic luggage (among all the other luggage) so that appropriate counter measures can be quickly taken (i.e. an operator can manually fix or adjust the position or configuration of the luggage). The solution should also be robust enough to accept feedback and learn and adapt to updating standards for identification of defective luggage.
Another problem is that an operator or a passenger, either accidentally or intentionally, enters an area, in which humans must not be allowed to enter (i.e. a keep-out zone), near an automatic conveyance system. This kind of incident may lead to a disruption of operation of the automatic conveyance system.
An object of the present invention is to provide a conveyor controlling apparatus and a conveyor controlling method have been created in light of such problems to improve the reliability of conveyance systems having conveyors.
A conveyor controlling apparatus for controlling a conveyance object on a conveyor in accordance with a conveyance state of the conveyance object, the apparatus comprises:
The conveyor controlling apparatus, wherein
The conveyor controlling apparatus, further comprises:
The conveyor controlling apparatus, wherein
The conveyor controlling apparatus further comprises:
The conveyor controlling apparatus further comprises:
The conveyor controlling apparatus, wherein
The conveyor controlling apparatus, wherein
The conveyor controlling apparatus further comprises:
A conveyor controlling method for controlling a conveyance object on a conveyor in accordance with a conveyance state of the conveyance object, the method comprises steps of:
The controlling method, further comprises steps of:
The controlling method further comprises steps of:
The controlling method, wherein
The controlling method further comprises steps of:
The controlling method further comprises steps of:
The controlling method, wherein
The controlling method further comprises a step of:
The controlling method further comprises a step of:
A conveyor controlling apparatus for controlling a conveyor in accordance with environment parameters with respect to the conveyor, the apparatus comprises:
The conveyor controlling apparatus, wherein
The conveyor controlling apparatus, wherein
The conveyor controlling apparatus, wherein
The conveyor controlling apparatus, wherein
The conveyor controlling apparatus, wherein
The conveyor controlling apparatus according to any one of claims 2 to 5, further comprises:
The conveyor controlling apparatus, further comprises:
The conveyor controlling apparatus, wherein
The conveyor controlling apparatus, further comprises:
A conveyor controlling method for controlling a conveyor in accordance with environment parameters with respect to the conveyor, the method comprises steps of:
The method, wherein
The method further comprises steps of:
The method, wherein
The method, further comprises steps of:
The method, further comprises a step of:
The method, further comprises steps of:
The method, further comprises steps of:
The method, wherein
The method, further comprises steps of:
The first embodiment of the present invention will now be described with reference to the accompanying main drawings
As shown in
That is, not only ordinary objects (e.g. bags or parcels), which are normally carried by the belt conveyor BC, but also other non-ordinary objects, which should theoretically not be carried by the belt conveyor BC, shall also be included in the definition of ‘conveyance object’ Cob.
The conveyor supervisor 1000 is designed for an airport AP. The airport AP has a non-restricted area (so-called Front of House (FOH)) and a restricted area (so-called Back of House (BOH)) as shown in
An automatic baggage handling system (BHS) Bap installed in the airport AP mainly has an introduction part Bi, a main part Bm, and a transition part Bt.
The introduction part Bi has a group of conveyors such as weighing conveyors (not shown in FIGs), holding conveyors (not shown in FIGs) and collector conveyors BC101 installed in first and second islands IS1 and IS2 provided in the FOH in the airport AP. Each of the first and second islands IS1 and IS2 has a plurality of self-baggage drop machines SBD.
Each self-baggage drop machine SBD has a weighing conveyor.
Each of the holding conveyors, which is provided at a downstream side of each of the weighing conveyors, receives checked-luggage (conveyance object Cob) forwarded from the weighing conveyor, and automatically drops the forwarded checked-luggage into the collector conveyor BC101 at an appropriate timing.
The collector conveyor BC101 is a belt conveyor that continuously runs substantially at a constant speed (e.g. 30 m/min) for conveying the checked-luggage dropped from the holding conveyor to the downstream-side.
The main part Bm is the core part of the automatic baggage handling system Bap provided in the BOH of the airport AP. The main part Bm has a group of conveyors such as main conveyors, tilt-tray conveyors, and make-up conveyors.
The transition part Bt, which is a part provided between the introduction part Bi and the main part Bm in the BOH of the airport AP, has a transport conveyor BC102 installed near an inlet of BOH.
The transport conveyor BC102, which is a belt conveyor disposed between a downstream end of the introduction part Bi and an upstream end of the main part Bm, forwards the checked-luggage (conveyance object Cob) sent from the collector conveyor BC101 to the main part Bm.
Each of the conveyors of the automatic baggage handling system Bap is controlled by a conveyor controller CTRBC (see
The conveyor supervisor 1000 is installed in a position abutting onto the collector conveyor BC101 (see
As shown in
The main body 101 is a pedestal portion of which a bottom face 116 is placed on a floor FL and vertically extends upwards. Further, as shown in
A top face 113 of the main body 101 is formed as a plane extending diagonally downward from a back face 112 to a front face 111. A touch screen (display unit) 305 is provided on the top face 113.
The pillar 102 is a metal member extending upward from the main body 101. A status indicator 306 is provided on a front face 117 of the pillar 102.
Further, the pillar 102 is slidably fixed to the main body 101 so that the height of the sensor head 103 is adjustable as shown by the bi-directional arrow Ash in
The sensor head 103 is an assembly unit fixed to the top end of the pillar 102. As shown in
Each of the left RGB camera 204 and the right RGB camera 206 is a Global Shutter RGB type camera with a fixed manual focus lens.
Further, as shown in
Each of the left three-dimensional camera 203 and the left RGB camera 204 captures an image of the conveyance object Cob as an upstream image (first image) when the conveyance object Cob is in an first upstream-assessment zone (upstream assessment zone) Z1u on the collector conveyor BC101 and while the conveyance object Cob is conveyed by the collector conveyor BC101 in a direction (see the arrow Ac in
The upstream image captured by the left three-dimensional camera 203 is transmitted in real time to the primary AI computer 301 as data processed three-dimensionally (3D point cloud data) about the conveyance object Cob. The upstream image captured by the left RGB camera 204 is simply forwarded in real time to the primary AI computer 301 as raw data that is not subjected to special processing.
As shown in
The downstream image captured by the right three-dimensional camera 205 is transmitted in real time to the primary AI computer 301 as data processed three-dimensionally (3D point cloud data) about the conveyance object Cob. The downstream image captured by the right RGB camera 206 is simply forwarded in real time to the primary AI computer 301 as raw data that is not subjected to special processing.
As shown in
In the present embodiment, the conveying direction Ac of the collector conveyor BC101 is the right direction (see
As shown in
The second downstream-assessment zone Z2d on the collector conveyor BC101 is another virtual area that approximately corresponds to a photographable area of the cameras towards the downstream direction Ad (i.e. the right three-dimensional camera 205 and the right RGB camera 206 in this embodiment).
Further, the first upstream-assessment zone Z1u is provided in the upstream direction Au (i.e. the left side in the present embodiment) with respect to the sensor head 103. On the other hand, the second downstream-assessment zone Z2d is provided in the downstream direction Ad (i.e. the right side in the present embodiment) with respect to the sensor head 103.
This means that the first upstream-assessment zone Z1u is set in the upstream direction Au with respect to the second downstream-assessment zone Z2d. The second downstream-assessment zone Z2d is set in the downstream direction Ad with respect to the first upstream-assessment zone Z1u.
Each of the first upstream-assessment zone Z1u and the second downstream-assessment zone Z2d is set as a stationary area regardless of running and stopping of the collector conveyor BC101.
A zone-separation distance (see a sign Dz in
The zone-separation distance Dz can be adjusted by changing the height of the sensor head 103, as shown by the arrow Ash in
In addition, the zone-separation distance Dz can be adjusted by individually changing mounting angle of the left three-dimensional camera 203, the left RGB camera 204, the right three-dimensional camera 205, and the right RGB camera 206, with respect to the sensor head 103.
For example, if the processing power of the primary AI computer 301 is relatively higher, then the zone-separation distance Dz may be set relatively shorter. On the other hand, if the processing power of the primary AI computer 301 is relatively lower, then the zone-separation distance Dz may be set relatively longer.
As shown in
The right light 208 is another lighting unit that illuminates the second downstream-assessment zone Z2d so that the right three-dimensional camera 205 and the right RGB camera 206 clearly capture the downstream image of the conveyance object Cob.
As shown in
The primary AI computer 301 is also communicably connected to each of the left three-dimensional camera 203, the left RGB camera 204, the right three-dimensional camera 205, and the right RGB camera 206. The AI computer 301 executes image-processing applications for image classification, object detection in images and division of images, in addition to the neural network processing.
Based on a learning model 307 stored in a storage device of the primary AI computer 301, the primary AI computer 301 analyses the upstream image and/or the downstream image, and recognises the ‘conveyance state’ with regard to the conveyance object Cob on the collector conveyor BC101. Further, the primary AI computer 301 determines whether the recognised conveyance state satisfies the ‘forwarding condition’ of the automatic baggage handling system Bap.
In the meantime, the determination on the conveyance state of the conveyance object Cob in the first upstream-assessment zone Z1 is referred to as the ‘primary determination’. On the other hand, the determination on the conveyance state of the conveyance object Cob in the second downstream-assessment zone Z2d is referred to as the ‘secondary determination’.
The learning model 307 defines an assessment standard. Specifically, the learning model 307 is a machine learning model generated based on a vast amount of raw data (training data), in which a number of conveyance objects Cob are photographed individually, and a learning data set corresponding to the factors of the conveyance objects Cob. The learning model 307 is stored in the storage device of the primary AI computer 301.
In addition, the primary AI computer 301 (training module) updates the learning model 307, in response to an overriding command containing an overriding reason entered by an operator via the touch screen 305 based on the entered overriding reason and the upstream image and/or the downstream image.
In the meantime, the ‘conveyance state’ includes various factors with regard to the conveyance object Cob (e.g. dimensions, orientation, type, condition of accessories, presence or absence of overlapping, distance from other conveyance objects Cob, presence or absence of transport tub, open/close state, presence or absence of baggage tag, and so on).
For example, when a conveyance object Cob is conveyed by the collector conveyor BC101, the primary AI computer 301, based on the learning model 307 and the upstream image and/or the downstream image, recognises the type of the conveyance object Cob as a suitcase (hard bags), a duffel bag (soft bags), a backpack (soft bags), a sports bag (soft bags), or others.
In addition, when a conveyance object Cob has wheels (movable parts), the primary AI computer 301 recognises whether the conveying direction Ac of the collector conveyor BC101 and a projecting direction of the wheels protrude from the conveyance object Cob are substantially the same. Further, the primary AI computer 301 recognises whether the orientation of the conveyance object Cob is either upright or lying flat. As shown in
Further, the primary AI computer 301 recognises whether an auxiliary piece such as a handle (movable parts) of a conveyance object Cob is appropriately stowed in the conveyance object Cob. In this regard,
In addition,
The primary AI computer 301 also recognises whether one conveyance object Cob overlaps another conveyance object Cob (see suitcases Cob4 and Cob5 in
In addition, the primary AI computer 301 recognises whether a distance (see the arrow D1 in
In addition, if a baggage tag is affixed to a conveyance object Cob, the primary AI computer 301 recognises the affixed baggage tag.
In addition, if there is relatively large damage to a conveyance object Cob, the primary AI computer 301 recognises the damage on the conveyance object Cob.
Further, the primary AI computer 301 recognises whether a conveyance object Cob is placed in a transport tub.
The primary AI computer 301 also recognises whether a shoulder strap is properly bundled if a conveyance object Cob has the shoulder strap.
The primary AI computer 301 also recognises whether a conveyance object Cob is properly closed (i.e. the open/close state of the conveyance object Cob).
That is, the primary AI computer 301 assesses a conveyance object Cob as an ‘improper-state object’, which is an object that is not in a state that can be conveyed, if the primary AI computer 301 recognises at least one of:
The primary AI computer 301 also assesses a conveyance object Cob as the ‘improper-state object’ if the primary AI computer 301 recognises that there are extraneous items (i.e. passports, books, magazines, documents and so on) left on the conveyance object Cob on the collector conveyor BC101.
In addition, the primary AI computer 301 assesses a conveyance object Cob on the collector conveyor BC101 as an ‘unknown object’, if the primary AI computer 301 cannot recognise the conveyance object Cob.
Further, the primary AI computer 301 determines that the conveyance state of the conveyance object Cob on the collector conveyor BC101 does not meets the forwarding condition, if the conveyance object Cob is assessed as at least one of the ‘improper-state object’ and the ‘unknown object’. On the other hand, the primary AI computer 301 determines that the conveyance state of the conveyance object Cob on the collector conveyor BC101 meets the forwarding condition, if the conveyance object Cob is assessed as neither the ‘improper-state object’ nor the ‘unknown object’.
The primary AI computer 301 and the system computer 303 ascertain the conveying direction Ac of the collector conveyor BC101 based on the installation information entered by an operator through the touch screen 305. Alternatively, the primary AI computer 301 and the system computer 303 may determine that the conveying direction Ac of the collector conveyor BC101 is either rightwards (see
As shown in
The system computer 303 is a computer that holistically control the conveyor supervisor 1000. The system computer 303 is communicably connected to each of the network hub 302, the I/O controller 304, and the touch screen 305.
The system computer 303 displays the processing status by the primary AI computer 301 on the touch screen 305. For instance,
When the primary AI computer 301 determines that the conveyance state of the conveyance object Cob does not meet the forwarding condition, then the system computer 303 displays the problems of the conveyance state on the touch screen 305. For example, an indication on the touch screen 305 is shown in
The system computer 303 has an external-communication module (remote accessing module) which is not shown in the FIGs. The system computer 303 is communicably connected to the Internet 904 via the external-communication module. This allows remote operators to connect to the system computer 303 via the Internet 904 and an external-communication module in order to remotely control the conveyor supervisor 1000.
The external-communication module (communication module) of the system computer 303 also communicates with other external systems (e.g. SCADA systems, systems under the control of the International Air Transport Association (IATA), and other cargo systems).
The system computer 303 has a diagnosing module (diagnosing module) that executes a diagnostic program for self-diagnosing the conveyor supervisor 1000.
The system computer 303 also has a logging module (logging module) that records the activity of the conveyor supervisor 1000.
In the meantime, the diagnostic module and the logging module are not shown in the FIGs.
The system computer 303 turns on the status indicator (alarm device) 306 via the I/O controller 304 in accordance with the results of the primary and secondary determinations performed by the primary AI computer 301. As a result, it is possible to visually notify operators of the conveyance state of the conveyance object Cob.
The touch screen 305 is an interface device provided on the top face 113 of the main body 101. The touch screen 305 is communicably connected to the system computer 303. The touch screen 305 visually notifies various information to an operator in accordance with the commands from the system computer 303. Further, the touch screen 305 receives inputs from the operator and forward the inputs entered by the operator to the system computer 303.
The I/O controller 304 is a control unit that is communicably connected to the system computer 303 and the conveyor controller CTRBC of the automatic baggage handling system Bap. Hence, the I/O controller 304 controls the collector conveyor BC101 via the conveyor controller CTRBC in accordance with control commands received from the system computer 303.
That is, the collector conveyor BC101 is controlled by the I/O controller 304, which works in accordance with the control commands transmitted by the system computer 303 according to the results of the primary and/or secondary determinations.
If the primary AI computer 301 determines that the conveyance state of the conveyance object Cob, which is recognised based on the learning model 307 and the upstream image, ‘satisfy’ (i.e. when the primary determination is a positive result), then the system computer 303 does not actively control over the I/O controller 304. Hence, in this case, the collector conveyor BC101 continues normal operation. As a result, the conveyance object Cob passes through the second downstream-assessment zone Z2d, and continues travelling in the downstream direction Ad.
On the other hand, if the primary AI computer 301 determines that the conveyance state of the conveyance object Cob, which is recognised based on the learning model 307 and the upstream image, ‘does not satisfy’ (i.e. when the primary determination is a negative result), the system computer 303, via the I/O controller 304 and the conveyor controller CTRBC, controls the collector conveyor BC101 in such a matter that the conveyance object Cob is stopped in the second downstream-assessment zone Z2d second downstream-assessment zone Z2d. This allows the right three-dimensional camera 205 and the right RGB camera 206 to capture the downstream image of the conveyance object Cob in a stationary state.
If the primary AI computer 301 determines that the conveyance state of the conveyance object Cob, which is recognised based on the learning model 307 and the downstream image, ‘does not satisfy’ (i.e. when the secondary determination is a negative result), the system computer 303 does not actively control over the I/O controller 304. Hence, in this case, the stationary state of the collector conveyor BC101 continues, so the conveyance object Cob is held as it is in the second downstream-assessment zone Z2d second downstream-assessment zone Z2d.
Further, if the secondary determination is a negative result, the system computer 303 executes a ‘need-of-care control’. This need-of-care control includes a control for indicating the problems of the conveyance state of the conveyance object Cob on the touch screen 305, and another control for turning on the status indicator 306. Details of this need-of-care control will be described later using
On the other hand, if the primary AI computer 301 determines that the conveyance state of the conveyance object Cob, which is recognised based on the learning model 307 and the downstream image, is ‘satisfactory’ (i.e. when the secondary determination is a positive result), then the system computer 303 sends a resume signal to the I/O controller 304. Afterwards, the I/O controller 304 restarts, via the conveyor controller CTRBC, the collector conveyor BC101 in the stationary state in response to the resume signal received from the system computer 303. As a result, the conveyance object Cob on the collector conveyor BC101 travels in the downstream direction Ad.
In the meantime, a situation in which the conveyance state of the conveyance object Cob does not satisfy the forwarding condition in the first upstream-assessment zone Z1u, but then satisfies the forwarding condition in the second downstream-assessment zone Z2d, it could generally be assumed that the conveyance condition of the conveyance object Cob has been corrected with an action taken by an operator because of executing the need-of-care control which enables the operator to recognise that the conveyance object Cob held in the second downstream-assessment zone Z2d does not meet the forwarding condition.
However, even if the primary AI computer 301 determines that the conveyance state of the conveyance object Cob does not meet the forwarding condition in the secondary determination, the I/O controller 304 receives the resume signal from the system computer (overriding module) 303 when the system computer 303 executes an ‘overriding operation’.
This overriding operation is an operation that is executed when an input of the overriding command to the touch screen 305 is completed. That is, when the input of the overriding command to the touch screen 305 is completed, the system computer 303 transmits the resume signal to the I/O controller 304. Then, the I/O controller 304, which received the resume signal, resumes the collector conveyor BC101 in the stationary state via the conveyor controller CTRBC so that the conveyance object Cob travels in the downstream direction Ad.
In other words, this overriding operation is carried out under a situation where the primary AI computer 301 determines that the conveyance state of the conveyance object Cob suspended in the second downstream-assessment zone Z2d still does not meet the forwarding condition, but an operator completed the input of the overriding command into the touch screen 305.
Then, when this overriding operation is executed, the system computer 303 restarts the collector conveyor BC101 via the I/O controller 304 and the conveyor controller CTRBC in order to send the conveyance object Cob in the downstream direction Ad.
Nevertheless, prior to executing the overriding operation (i.e. for completion of inputting the overriding command), it is required to input a reason (overriding reason) into the touch screen 305.
As shown in
For example, assuming that the primary AI computer 301 recognises that the size of a conveyance object Cob held in the second downstream-assessment zone Z2d exceeds the predetermined specified size (height, width and depth), as a result, the primary AI computer 301 determines that the conveyance object Cob is deemed as an improper-state object, and therefore the primary AI computer 301 determines that the conveyance state of the conveyance object Cob does not satisfy the forwarding condition, but the operator has confirmed that the size of the conveyance object Cob stopped in the second downstream-assessment zone Z2d is actually within the predetermined specified dimensions.
In this case, the operator enters a tick into the check box of ‘Acceptable size’, which is one of the overriding reasons displayed on the touch screen 305 as shown in
When the overriding operation is executed, the system computer 303 forwards the overriding reason entered into the touch screen 305 to the primary AI computer 301. The primary AI computer 301, which received this overriding reason, updates the learning model 307 based on the upstream and downstream images corresponding to the conveyance object Cob subjected to the overriding operation and the overriding reason received from the system computer 303.
The states of the learning model 307 can be restored as needed. For example, backup data of the learning model 307 may be periodically saved and stored in the storage device of the primary AI computer 301. This affords the ability to restore the learning model 307 based on the backup data of any restore points. This is particularly useful in the event of any ‘incorrect training data’ which had been fed into and has corrupted the learning model 307.
Alternatively, more simply, a factory default version of backup data of the learning model 307 may be stored in the storage device of the primary AI computer 301. According to this arrangement, the learning model 307 can be restored to the factory default state at any time.
In other words, even if operators repeatedly enter the overriding reason incorrectly, the learning model 307 can be corrected by restoring the learning model 307 as appropriate.
The collector conveyor BC101 has a first photo-eye sensor 209 near the first upstream-assessment zone Z1u. Further, the collector conveyor BC101 has a second photo-eye sensor 211 near the second downstream-assessment zone Z2d.
Each of the first and second photo-eye sensors 209 and 211 is communicably connected to the I/O controller 304.
Hence, the system computer 303 and the primary AI computer 301 may detect the conveyance object Cob in the first upstream-assessment zone Z1u by the first photo-eye sensor 209.
Likewise, the system computer 303 and the primary AI computer 301 may detect the conveyance object Cob in the second downstream-assessment zone Z2d by the second photo-eye sensor 211.
With the above-mentioned arrangement, the conveyor controlling apparatus according to the first embodiment of the present invention provides the following advantages.
The assessment of the conveyance object Cob by the conveyor supervisor 1000 is mainly performed in accordance with a main flowchart shown in
At step S001, when a conveyance object Cob, which is conveyed in the downstream direction Ad by the collector conveyor BC101, enters into the first upstream-assessment zone Z1u, then the left light 207 illuminates the first upstream-assessment zone Z1u and the left three-dimensional camera 203 and the left RGB camera 204 capture an image of the conveyance object Cob as the upstream image.
Then, the primary AI computer 301 performs the image recognition using the learning model 307 for the captured upstream image.
Afterwards, at step S005, the primary AI computer 301 determines whether the conveyance state of the conveyance object Cob on the collector conveyor BC101 satisfies the forwarding condition based on the recognition result of the upstream image.
If the result of the determination at step S005 is positive (see Yes route from step S005), the system computer 303 does not actively control the conveyor controller CTRBC. That is, in this case, the collector conveyor BC101 continues normal operation. Accordingly, the conveyance object Cob on the collector conveyor BC101 passes through the second downstream-assessment zone Z2d and is further carried in the downstream direction Ad.
On the other hand, if the result of the decision at step S005 is negative (see No route from step S005), the system computer 303 instructs the conveyor controller CTRBC so that the conveyance object Cob on the collector conveyor BC101 is stopped within the second downstream-assessment zone Z2d (step S010).
At step S015, an image of the conveyance object Cob stationary in the second downstream-assessment zone Z2d is captured as the downstream image by the right three-dimensional camera 205 and the right RGB camera 206. At this time, the right light 208 illuminates the second downstream-assessment zone Z2d.
Afterwards, the primary AI computer 301 performs the image recognition using the learning model 307 for the captured downstream image.
Thereafter, at step S020, the primary AI computer 301 determines whether the conveyance state of the conveyance object Cob on the collector conveyor BC101 satisfies the forwarding condition based on the recognition result of the downstream image.
If the determination result in this the step S020 is positive (see Yes route from step S020), the system computer 303 sends the resume signal to the I/O controller 304. The I/O controller 304 received the resume signal sends the resume request to the conveyor controller CTRBC. The conveyor controller CTRBC received this resume request restarts the collector conveyor BC101 in the stationary state (step S025). As a result, the conveyance object Cob on is conveyed further in the downstream direction Ad.
On the other hand, if the determination at step S020 is negative (see No route from step S020), the system computer 303 does not actively control the I/O controller 304 and the conveyor controller CTRBC. That is, in this case, the stationary state of the collector conveyor BC101 continues. As a result, the conveyance object Cob on the collector conveyor BC101 is kept holding as it is in the second downstream-assessment zone Z2d.
At this time, at step S030, the system computer 303 executes the ‘need-of-care control’. In this need-of-care control, as shown in
Specifically, at step S045, the system computer 303 indicates the problems of the conveyance state of the conveyance object Cob on the touch screen 305 (see
At step S050, the system computer 303 turns on the status indicator 306 via the I/O controller 304. This visually provides a call to an operator's attention even if he/she is a short distance away from the conveyor supervisor 1000. That is, it is possible to notify the operator that the conveyance object Cob held in the second downstream-assessment zone Z2d is requiring some assistance.
Afterwards, at step S035 of
On the other hand, when the input of the overriding command has not been completed (see No route from step S035), the conveyance object Cob on the collector conveyor BC1 is held as it is at the second downstream-assessment zone Z2d.
Thereafter, when the input of the overriding command is completed and the overriding operation is executed (see Yes route from step S035), then at step S040, the system computer 303 forwards the overriding reason entered via the touch screen 305 to the primary AI computer 301. The primary AI computer 301 updates the learning model 307 based on the upstream and downstream images of the conveyance object Cob, which was subject to the overriding operation, and the overriding reason received from the system computer 303.
Then, at step S025, the system computer 303 transmits the resume signal to the I/O controller 304. The I/O controller 304 received the resume signal, via the conveyor controller CTRBC, restarts the collector conveyor BC101 in the stationary state. As a result, the conveyance object Cob is further carried in the downstream direction Ad.
The present invention is not limited to the above-described embodiment and the variant thereof, and can be variously modified and implemented.
As shown in
The CCTV images captured by the CCTV camera 212 are recorded and saved to a data storage server SSVR which is communicably connected to the network hub 302 of the conveyor supervisor 1000.
The CCTV images captured by the CCTV camera 212 are processed by a secondary AI computer 308 which is installed in the main body 101 of the conveyor supervisor 1000. The secondary AI computer 308 is a single-board computer that is capable of neural network processing based on image data. The secondary AI computer 308 is communicably connected to the network hub 302.
The secondary AI computer 308 is used to detect a suspicious person (e.g. a person who has intruded into an area near the conveyor supervisor 1000 or on the collector conveyor BC101 and a child who has accidentally strayed on the collector conveyor BC101) by analysing and recognising people captured in the CCTV image, based on a CCTV learning model 308 stored in the storage device of the secondary AI computer 308 and the CCTV images taken by the CCTV camera 212.
The analysis and recognition results of the CCTV images obtained by the secondary AI computer 308 are stored in the data storage server SSVR.
Accordingly, the recognition result of the CCTV images and the CCTV images stored on the data storage server SSVR can be checked as necessary.
Further, when the secondary AI computer 308 detects a suspicious person based on the CCTV images, the system computer 303 may urgently stop the collector conveyor BC101 via the I/O controller 304 and the conveyor control unit CTRBC.
Also, when the secondary AI computer 308 detects a suspicious person, the system computer 303 may send a signal, which is for notifying that the suspicious person has been detected, to other external systems (e.g. a security system of the airport).
The primary AI computer 301 may determine the conveying direction Ac of the collector conveyor BC101 based on the upstream image and/or the downstream image. That is, the primary AI computer 301 judges the conveying direction Ac of the collector conveyor BC101 is either rightwards (see
In the above embodiment, if the result of the determination at the step S005 shown in
For example, as shown in
According to this arrangement, the learning model 307 can be further trained based on a large number of the downstream images.
As shown in
That is, in the configuration as shown in
Accordingly, even when the conveyance object Cob is stopped in the second downstream-assessment zone Z2d, it is possible to suppress an impact onto handling process carried out by the automatic baggage handling system Bap.
As described in detail above, according to the conveyor controlling apparatus and method of the present invention, it is possible to assess appropriately whether the conveyance state of various types of conveyance objects meets different forwarding conditions for each automatic conveyance system so that the conveyance objects are precisely controlled in accordance with the assessed conveyance state.
Specifically, according to the conveyor controlling apparatus and method of the present invention, an operator can be swiftly made aware of a conveyance object, which is halted in the second downstream-assessment zone having some problems that does not meet the forwarding condition.
Further, it is also possible to quickly and automatically resume temporarily halted forwarding of the conveyance object in the downstream direction once the problems of the conveyance object are fixed by the operator as the conveyor is automatically restarted when satisfying of the forwarding condition is determined based on the downstream image.
In addition, it is possible to improve the accuracy of determination on the conveyance state by recognising the conveyance state of the conveyance object based on the learning model and the upstream image and/or the downstream image, and determining whether the recognised conveyance state satisfies the forwarding condition.
In addition, it is possible to improve automatically the accuracy of the learning model by simply carrying out the normal handling procedures for the conveyance object.
That is, an operator is simply required to input the reason why the operator needs to execute the overriding operation, but in fact, the learning model is updated based on this reason (i.e. the overriding reason) and the upstream and/or downstream images. Accordingly, the accuracy of the learning model can be automatically improved without requiring educating operators on how to update the learning model.
In addition, it is possible to easily and reliably capture the upstream and downstream images by arranging the upstream capturing device and the downstream capturing device being adjacent to one another and orientated in opposing directions.
Further, it is possible to capture clearly the upstream image and the downstream image of the conveyance object from above by mounting the upstream capturing device and the downstream capturing device in the head unit.
Further, it is possible to capture clearly the upstream image and the downstream image by illuminating the first upstream-assessment zone and the second downstream-assessment zone.
In addition, it is possible to capture easily and clearly the upstream and downstream-side images by providing the upstream and downstream-side lighting devices at the head unit.
In addition, the zone-separation distance is adjustable according to the processing capability related to the assessment of the conveyance state.
For example, if the processing power of the image-analysing module is relatively higher, then the zone-separation distance may beset relatively shorter. On the other hand, if the processing power of the image-analysing module is relatively lower, then the zone-separation distance may be set relatively longer.
Specifically, in the above-described first embodiment, the collector conveyor BC101 continues its operation while the primary AI computer 301 determines whether the conveyance state of the conveyance object Cob on the collector conveyor BC101 satisfies the forwarding condition based on the learning model 307 and the upstream image (i.e. while the primary determination is carried out). As a result, the conveyance object Cob continues to move in the downstream direction Ad even while the processing of the primary determination is ongoing. However, in this case, the right three-dimensional camera 205 and the right RGB camera 206 could not capture the conveyance object Cob, if a result of the primary determination came out after the conveyance object Cob has passed the second downstream-assessment zone Z2d and further moved in the downstream direction Ad.
For this reason, the zone distance is set taking into account the processing power of the image-analysing module (primary AI computer 301).
In addition, it is possible to seek quickly and reliably a possible assistance from an operator by visually and/or aurally notifying an operator of a result of the determination whether the conveyance state of the conveyance object meets the forwarding condition.
This also allows the operator to do other work until the operator gets a visual and/or aural notification that the forwarding condition has not been satisfied. Accordingly, the work efficiency of the operator may be improved.
The second embodiment of the present invention will now be described using mainly
As shown in
The conveyor supervisor 2000 is designed for an airport AP in the same manner of the conveyor supervisor 1000 according to the first embodiment. The conveyor supervisor 2000 can also be utilised in both the FOH and the BOH of the airport AP. In this airport AP, each of the conveyor supervisors 2000 is installed at the points shown as “Baz1”, “Baz2”, and “Baz3” in
The conveyor supervisor 2000 is installed in a position abutting onto the collector conveyor BC101 (see
As shown in
The main body 2101 is a pedestal portion of which a bottom face 2116 is placed on a floor FL and vertically extends upwards. Further, as shown in
A DC power supply unit is also provided in the main body 2101, but illustration is omitted.
A top face 2113 of the main body 2101 is formed as a plane extending diagonally downward from a back face 2112 to a front face 2111. A touch screen (display unit) 2305 is provided on the top face 2113.
The pillar 2102 is a member extending upward from the main body 2101. A status indicator 2306 is provided on a front face 2117 of the pillar 2102.
Further, the pillar 2102 is slidably fixed to the main body 2101 so that the height of the sensor head 2103 is adjustable as shown by the bi-directional arrow Ash in
The sensor head 2103 is an assembly unit fixed to the top end of the pillar 2102. As shown in
A CCTV camera (intrusion camera) 2309 is a security camera and its frame rate can be adjusted.
The images (intrusion image) captured by the CCTV camera 2309 are transmitted to the secondary computer 2308 via the Ethernet switch 2302 as image stream data.
The CCTV camera 2309 is adjustably mounted to the base-plate 2113 via a ball-joint 2212M1 and a swing-shaft 2212M2. In this example shown in
Further, as shown in
Each of the left three-dimensional camera 2203 and the left RGB camera 2204 captures an image of the conveyance object Cob as a ‘first image’ when the conveyance object Cob is in an first upstream-assessment zone Z1u on the collector conveyor BC101 and while the conveyance object Cob is conveyed by the collector conveyor BC101 in a direction (see the arrow Ac in
In addition, each of the left three-dimensional camera 2203 and the left RGB camera 2204 captures an image of the conveyance object Cob stationary in a second upstream-assessment zone Z2u on the collector conveyor BC101 as a ‘second image’.
Each of the images captured by the left three-dimensional camera 2203 is transmitted in real time to the primary computer 2301 as data processed three-dimensionally (3D point cloud data) about the conveyance object Cob. Each of the images of the conveyance object Cob captured by the left RGB camera 2204 is transmitted in real time to the primary computer 2301 as raw data that is not subjected to special processing.
In the present embodiment, the conveying direction Ac of the collector conveyor BC101 is the right direction (see
The left RGB camera 2204 is a Global Shutter RGB type camera with a fixed manual focus lens which is directed to the assessment zone Za including both of the first upstream-assessment zone Z1u and the second upstream-assessment zone Z2u.
The left RGB camera 2204 is communicably connected to the primary computer 2301 with a USB3 or similar cable.
The image data transmitted from the left RGB camera 2204 is forwarded by the primary computer 2301 to the AI vision processor 2311 for the AI analysis carried out by the AI vision processor 2311.
The AI vision processor 2311 is a computer dedicated for neural net artificial intelligence (AI) processing.
The AI vision processor 2311 has a memory in which an image-analysing module 2401 and a learning model 2405 are stored.
The AI vision processor 2311 receives the data of the images captured by each of the left three-dimensional camera 2203 and the left RGB camera 2204 from the primary computer 2301.
The image-analysing module 2401, which is a software module, implements the learning model 2405 which has been preliminarily trained.
The image-analysing module 2401 assesses and determines whether the conveyance state of the conveyance object Cob satisfies the forwarding condition based on the received image data and the trained learning model 2405.
The learning model 2405 defines an assessment standard. Specifically, the learning model 2405 is a machine learning model generated based on a vast amount of raw data (training data), in which a number of conveyance objects are photographed individually, and a learning data set corresponding to the factors of the conveyance objects.
The AI vision processor 2311 transmits results of the assessment and determination process to the primary computer 2301.
The training module 2404 of the primary computer 2301 updates the learning model 2405, when the overriding operation (described later) is executed, based on the overriding reason entered into the touch screen 2305 and the image corresponding to the conveyance object Cob subjected to the overriding operation.
The training module 2404 restores the states of the learning model 2405 as needed.
For example, backup data of the learning model 2405 may be periodically saved and stored in the storage device of the primary computer 2301. This affords the ability to restore the learning model 2405 based on the backup data of any restore points. This is particularly useful in the event of any ‘incorrect training data’ which has been fed into and has corrupted the learning model 2405.
Alternatively, a factory default version of backup data of the learning model 2405 may be stored in the storage device of the primary computer 2301. According to this arrangement, the learning model 2405 can be restored to the factory default state at any time.
In other words, even if operators repeatedly enter the overriding reason incorrectly, the training module 2404 can correct the learning model 2405 by restoring the training module 2404 as appropriate.
As shown in
The assessment zone Za has the first upstream-assessment zone Z1u and the second upstream-assessment zone Z2u.
The assessment zone Za is provided in the upstream direction Au (i.e. the left side in the present embodiment) with respect to the sensor head 2103 shown in
Each of the first upstream-assessment zone Z1u and the second upstream-assessment zone Z2u is set as a stationary area regardless of running and stopping of the collector conveyor BC101.
The first upstream-assessment zone Z1u is provided in the upstream direction Au with respect to the second upstream-assessment zone Z2u.
The first upstream-assessment zone Z1u is an area in which an image of the conveyance object Cob (i.e. object image) is captured as a ‘first image’ by each of the left three-dimensional camera 2203 and the left RGB camera 2204 while the conveyance object Cob is conveyed by the collector conveyor BC101.
As schematically shown in
The first photo-eye sensor 2209 is an optical sensor which detects that the conveyance object Cob has entered the first upstream-assessment zone Z1u.
The first photo-eye sensor 2209 is connected to the I/O controller 2304 with a digital I/O cable.
Further, the I/O controller 2304 is communicably connected to the primary computer 2301 via the Ethernet switch 2302.
Hence, the primary computer 2301 can immediately detect that the conveyance object Cob enters into the first upstream-assessment zone Z1u by the first photo-eye sensor 2209.
The second upstream-assessment zone Z2u is provided in the downstream direction Ad with respect to the first upstream-assessment zone Z1u.
The second upstream-assessment zone Z2u is an area in which the object image is captured as a ‘second image’ by each of the left three-dimensional camera 2203 and the left RGB camera 2204 when the conveyance object Cob is stationary on the collector conveyor BC101 since the collector conveyor BC101 is stopped by the primary computer 2301.
Further, in this second upstream assessment zone Z2u, an operator can attend on the conveyance object Cob stationary on the collector conveyors BC101.
A second photo-eye sensor 2211 is provided beside an upstream edge of the second upstream-assessment zone Z2u.
The second photo-eye sensor 2211 is an optical sensor detecting that the conveyance object Cob has entered the second upstream-assessment zone Z2u.
The second photo-eye sensor 2211 is connected to the I/O controller 2304 with a digital I/O cable.
Further, the I/O controller 2304 is communicably connected to the primary computer 2301 via the Ethernet switch 2302.
Hence, the primary computer 2301 can immediately detect that the conveyance object Cob enters the second upstream-assessment zone Z2u by the second photo-eye sensor 2211.
A length of the first upstream-assessment zone Z1u (i.e. a first-zone length—see a sign ‘LZ1’ in
The first-zone length LZ1 can be adjusted by changing the height of the sensor head 2103, as shown by the arrow Ash in
In addition, the first-zone length LZ1 can be adjusted by individually changing mounting angle of the left three-dimensional camera 2203 and the left RGB camera 2204 with respect to the sensor head 2103.
As an example, if the processing power of the primary computer 2301 and the AI vision processor 2311 is relatively higher, then the first-zone length LZ1 may be set relatively shorter with considering the length of the assessment zone Za and the length of the second upstream-assessment zone Z2u. On the other hand, if the processing power of the primary computer 2301 and the AI vision processor 2311 is relatively lower, then the first-zone length LZ1 may be set relatively longer with considering the length of the assessment zone Za and the length of the second upstream-assessment zone Z2u.
As another example, if the running speed of the collector conveyor BC101 is relatively slower, then the first-zone length LZ1 may be set relatively shorter. On the other hand, if the running speed of the collector conveyor BC101 is relatively faster, then the first-zone length LZ1 may be set relatively longer.
As shown in
As shown in
The primary computer 2301 is connected to the Ethernet switch 2302 with an Ethernet cable.
Hence, the primary computer 2301 is communicably connected to the I/O controller 2304 via the Ethernet switch 2302.
The primary computer 2301 is connected to each of the left three-dimensional camera 2203 and the left RGB camera 2204 with a USB3 or similar cable.
The primary computer 2301 is connected to the touch screen 2305 with a HDMI cable.
The primary computer 2301 is communicably connected to the Internet as well as other external systems.
The primary computer 2301 has a storage device in which software modules such as a conveyer-controlling module 2402, an overriding module 2403 and a training module 2404 are stored.
The conveyer-controlling module 2402 stops the collector conveyor BC101 via a conveyor controller (not shown in FIGs) of the automatic baggage handling system Bap such that the conveyance object Cob is positioned stationary within the second upstream-assessment zone Z2u if the image-analysing module 2401 of the AI vision processor 2311 determines based on the first image that the conveyance state of the conveyance object Cob does not satisfy the forwarding condition.
Further, via the conveyor controller of the automatic baggage handling system Bap, the conveyer-controlling module 2402 restarts the collector conveyor BC101, which is stopped by the conveyer-controlling module 2402, if the image-analysing module 2401 determines based on the second image that the conveyance state of the conveyance object Cob satisfies the forwarding condition.
In the meantime, the determination on the conveyance state of the conveyance object Cob in the first upstream-assessment zone Z1u is referred to as the ‘primary determination’. On the other hand, the determination on the conveyance state of the conveyance object Cob in the second upstream-assessment zone Z2u is referred to as the ‘secondary determination’.
The overriding module 2403 executes an ‘overriding control’ that forcibly restarts the collector conveyor BC101 from the stationary state via the conveyor controller of the automatic baggage handling system Bap. This overriding operation is an operation that is run when an input of the overriding command to the touch screen 2305 is completed.
That is, when the input of the overriding command to the touch screen 2305 is completed, the overriding module 2403 executes the overriding control.
In other words, this overriding operation is carried out in a situation where the image-analysing module 2401 of the AI vision processor 2311 determined that the conveyance state of the conveyance object Cob did not satisfy the forwarding condition, but an operator completed inputting the overriding command into the touch screen 2305.
As shown in
For example, assuming that the image-analysing module 2401 of the AI vision processor 2311 recognises that the size of a conveyance object Cob exceeds the predetermined specified size (height, width and depth), as a result, the image-analysing module 2401 determines that the conveyance state of this conveyance object Cob does not satisfy the forwarding condition, but the operator has confirmed that the size of this conveyance object Cob stopped is actually within the predetermined specified dimensions.
In this case, the operator enters a tick into the check box of ‘Acceptable size’, which is one of the overriding reasons displayed on the touch screen 2305 as shown in
The training module 2404 updates the learning model learning model 2405, in response to an overriding command containing an overriding reason entered by an operator via the touch screen 2305 based on the entered overriding reason and the first image and/or the second image.
The training module 2404 sequentially records the results of the primary and secondary determinations by the conveyor control unit 2402 and the first and second images corresponding to these determinations into a storage server (not shown in FIGs). This storage server is an external server that is communicably connected to the conveyor supervisor 2000.
The ‘conveyance state’ includes various factors with regard to the conveyance object Cob (e.g. dimensions, orientation, type, condition of accessories, presence or absence of overlapping, distance from other conveyance objects Cob, presence or absence of transport tub, open/close state, presence or absence of baggage tag, and so on).
For example, when a conveyance object Cob is conveyed by the collector conveyor BC101, the image-analysing module 2401 of the AI vision processor 2311, based on the learning model 2405 and the first image and/or the second image, recognises the type of the conveyance object Cob as a suitcase (hard bags), a duffel bag (soft bags), a backpack (soft bags), a sports bag (soft bags), or others.
In addition, when a conveyance object Cob has wheels (movable parts), the image-analysing module 2401 of the AI vision processor 2311 recognises whether the conveying direction Ac of the collector conveyor BC101 and a projecting direction of the wheels protrude from the conveyance object Cob is substantially the same. Further, the image-analysing module 2401 recognises whether the orientation of the conveyance object Cob is either upright or lying. As shown in
Further, the image-analysing module 2401 of the AI vision processor 2311 recognises whether an auxiliary piece such as a handle (movable parts) of a conveyance object Cob is appropriately stowed in the conveyance object Cob. In this regard,
In addition,
The image-analysing module 2401 also recognises whether one conveyance object Cob overlaps another conveyance object Cob (see suitcases Cob4 and Cob5 in
In addition, the image-analysing module 2401 recognises whether a distance (see the arrow D1 in
If a baggage tag is affixed to a conveyance object Cob, the image-analysing module 2401 may recognise the affixed baggage tag.
If there is relatively large damage to a conveyance object Cob, the image-analysing module 2401 may recognise the damage on the conveyance object Cob.
Further, the image-analysing module 2401 recognises whether a conveyance object Cob is placed in a transport tub.
The image-analysing module 2401 also recognises whether a shoulder strap is properly bundled/stowed if a conveyance object Cob has the shoulder strap.
The image-analysing module 2401 may recognise whether a conveyance object Cob is properly closed (i.e. the open/close state of the conveyance object Cob).
That is, for example, the image-analysing module 2401 of the AI vision processor 2311 assesses that a conveyance state of a conveyance object Cob does not satisfy the forwarding condition (i.e. the conveyance state corresponds at least one of the rejection categories), if the image-analysing module 2401 recognises at least one of:
The conveyer-controlling module 2402 of the primary computer 2301 displays the processing status by the image-analysing module 2401 of the AI vision processor 2311 on the touch screen 2305. For instance,
When the forwarding condition is not met by the conveyance state of the conveyance object Cob, then the conveyer-controlling module 2402 of the primary computer 2301 displays the problems of the conveyance state on the touch screen 2305. For example, an indication on the touch screen 2305 is shown in
As shown in
The touch screen 2305 visually notifies various information to an operator in accordance with the commands from the touch screen 2305. Further, the touch screen 2305 receives inputs from the operator and forward the inputs entered by the operator to the touch screen 2305.
The primary computer 2301 ascertains the conveying direction Ac of the collector conveyor BC101 based on the installation information entered by an operator through the touch screen 2305. Alternatively, the primary computer 2301 may determine that the conveying direction Ac of the collector conveyor BC101 is either rightwards or leftwards based on the first image and/or the second image.
As shown in
The status indicator 2306 is an LED light unit that illuminates red, green, or amber for easy identification of the operational state of the conveyor supervisor 2000.
Specifically, the conveyer-controlling module 2402 of the primary computer 2301 controls the status indicator 2306 via the I/O controller 2304 in accordance with the working state of the conveyor supervisor 2000 as summarised below.
According to this, it is possible to visually notify operators of the conveyance state of the conveyance object Cob.
The I/O controller 2304 is a control unit that is communicably connected with a digital I/O cable to each of the first photo-eye sensor 2209, the second photo-eye sensor 2211, the status indicator 2306, the left light 2207, and a conveyor controller (not shown in FIGs) of the automatic baggage handling system Bap.
The I/O controller 2304 is also connected to the Ethernet switch 2302 with an Ethernet cable. Hence, the primary computer 2301 controls the collector conveyor BC101 via the I/O controller 2304.
The secondary computer 2308 is a supplementary computer that controls the processing related to an ‘intrusion control’ among the processes of the conveyor supervisor 2000.
The secondary computer 2308 is connected to the Ethernet switch 2302 with an Ethernet cable.
The secondary computer 2308 has a storage device in which software modules such as an intrusion-analysing module 2407 are stored.
The intrusion-analysing module 2407 implements an intrusion learning model (not shown in FIGs) which has been preliminarily trained for the intrusion control.
The secondary computer 2308 continuously receives image stream data fed from the CCTV camera 2309 via the Ethernet switch 2302.
The secondary computer 2308 runs dedicated human detection application algorithms implemented in the intrusion-analysing module 2407.
The intrusion-analysing module 2407 is designed for detecting the presence of humans in a region of interest which corresponds to a ‘keep-out zone’. The region of interest is calibrated to the collector conveyor BC101 in view of the image stream captured by the CCTV camera 2309.
The keep-out zone is an area extending along the collector conveyor BC101 and including at least the collector conveyor BC101.
For example, as shown in
In this case, the intrusion-analysing module 2407 of the secondary computer 2308 detects a presence of the infant Hi within the keep-out zone ZK as shown as shown in
In response to the detection of the infant Hinf, the intrusion-analysing module 2407 of the secondary computer 2308 immediately stops the collector conveyor BC101 via the conveyer-controlling module 2402 of the primary computer 2301.
With the above-mentioned arrangement, the conveyor controlling apparatus according to the second embodiment of the present invention provides the following advantages.
The analysis control of the conveyance object Cob by the conveyor supervisor 2000 is mainly performed in accordance with a main flowchart shown in
Note that, in parallel with the analysis control of the conveyance object Cob shown in
As shown in
Immediately subsequent to this, the conveyer-controlling module 2402 stays on standby to make a very short delay for ensuring that the entire conveyance object Cob enters the first upstream-assessment zone Z1u (step S210).
At this time, the conveyer-controlling module 2402 of the primary computer 2301 obtains the first image of the conveyance object Cob3 captured by each of the left three-dimensional camera 2203 and the left RGB camera 2204 (step 215). During this time, the collector conveyor BC101 is still running, hence, the conveyance object Cob3 keeps travelling in the downstream direction Ad.
Then, the image-analysing module 2401 of the AI vision processor 2311 determines whether the conveyance state of the conveyance object Cob3 satisfies the forwarding condition based on the first image captured by each of the left three-dimensional camera 2203 and the left RGB camera 2204 (steps S215 and S220 in
Also, at this time, the training module 2404 records the analysis results of the conveyance object Cob3 carried out by the image-analysing module 2401 of the AI vision processor 2311 and the captured first image of the conveyance object Cob3 in the external storage server (steps S215 and S220 in
If the conveyance state of the conveyance object Cob3 satisfies the forwarding condition (Yes route from step S220 in
On the other hand, if the conveyance state of the conveyance object Cob3 does not satisfy the forwarding condition (No route from step S220 in
The conveyer-controlling module 2402 recognises that the conveyance object Cob3 has entered the second upstream-assessment zone Z2u in response to the second photo-eye sensor 2211 detecting the conveyance object Cob3.
Nevertheless, the conveyer-controlling module 2402 is set to make a very short delay to ensure that the entire conveyance object Cob3 enters the second upstream-assessment zone Z2u (step S225 in
After that, at step S235 of
At step S400, the conveyer-controlling module 2402 indicates on the touch screen 2305 the determined problems with respect to the conveyance state of the conveyance object Cob3 (see
Further, at the step S405 shown in
According to this arrangement, even if an operator is a short distance from the conveyor supervisor 2000, the operator can be visually alerted. In other words, it is possible to swiftly notify the operator that some assistance is required for the conveyance object Cob3 stopped in the second upstream-assessment zone Z2u.
As a result, the problems of the conveyance object Cob3 are solved (step S240 in
At this time, if the operation mode of the conveyor supervisor 2000 with regard to restarting the collector conveyor BC101 is ‘automatic’ (Automatic route from step S245), the conveyer-controlling module 2402 obtains the second image of the conveyance object Cob3 captured by each of the left three-dimensional camera 2203 and the left RGB camera 2204 (step S270 in
Further, the image-analysing module 2401 of the AI vision processor 2311 determines whether the conveyance object Cob3 satisfies the forwarding condition based on the second image captured by the left three-dimensional camera 2203 and the left RGB camera 2204 (steps S270 and S275 in
Also, at this time, the training module 2404 records the analysis results of the conveyance object Cob3 carried out by the image-analysing module 2401 of the AI vision processor 2311, the determination results as to whether or not the conveyance state of the conveyance object Cob3 satisfies the forwarding condition, and the captured second image of the conveyance object Cob3 in the external storage server (steps S270 and S275 of
If the conveyance state of the conveyance object Cob3 satisfies the forwarding condition (Yes route from S275), the conveyer-controlling module 2402 restarts the collector conveyor BC101 via the conveyor controller of the automatic baggage handling system Bap (step S280). At this time, in the example shown in
On the other hand, if the operation mode of the conveyor supervisor 2000 with regard to restarting the collector conveyor BC101 is ‘manual’ (Manual route from step S245), the conveyer-controlling module 2402 indicates a ‘Force-in’ button and a ‘Retry’ button on the touch screen 2305.
If the Force-in button is touched by an operator (see Force-in route from step S250 in
In other words, this situation could be, for example, that although an operator has checked the conveyance object Cob3 at step S240, no problem was actually found, and therefore the operator believes that the conveyance object Cob3 should continue travelling in the downstream direction Ad.
At this time, the operator appropriately checks the check boxes corresponding to the reasons why there is no problem with the conveyance object Cob3 (i.e. ‘overriding reason’), then the operator touches a ‘Submit’ button indicated on the touch screen 2305 (step S255).
Also, at this time, the training module 2404 records the analysis results of the conveyance object Cob3 carried out by the image-analysing module 2401 of the AI vision processor 2311, the determination results as to whether or not the conveyance state of the conveyance object Cob3 satisfies the forwarding condition, the captured first image of the conveyance object Cob3, and the overriding reason entered through the touch screen 2305 in the external storage server (step S260).
In addition, the training module 2404 may apply a learning control to the learning model 2405 based on the entered overriding reason.
In short, in a situation where the image-analysing module 2401 of the AI vision processor 2311 has determined that the conveyance state of the conveyance object Cob3 does not satisfy the forwarding condition, but an operator has completed inputting the overriding command through the touch screen 2305, therefore, the overriding module 2403 executes the overriding control to forcibly restart the collector conveyor BC101 via the conveyor controller of the automatic baggage handling system Bap (step S265 of
In contrast, at the step S250, if an operator touches the Retry button (see Retry route from step S250), the conveyer-controlling module 2402 obtains the second image of the conveyance object Cob3 captured by each of the left three-dimensional camera 2203 and the left RGB camera 2204.
In other words, this situation could be, for example, that although an operator has checked the conveyance object Cob3 at the step S240, no problem was actually found, and therefore the operator wants to have the conveyor supervisor 2000 reconfirm the conveyance state of the conveyance object Cob, as a precaution.
As another example situation, an operator has solved the problems of the conveyance object Cob3 at the step S240, and as a precaution, the operator wants to make sure that the correction to the conveyance object Cob3 was appropriate by using the conveyor supervisor 2000.
If the Retry button is touched (see Retry route from step S250), the image-analysing module 2401 of the AI vision processor 2311 determines, based on the second image captured by each of the left three-dimensional camera 2203 and the left RGB camera 2204, whether or not the conveyance state of the conveyance object Cob3 satisfies the forwarding condition (steps S270 and S275).
If the conveyance state of the conveyance object Cob3 satisfies the forwarding condition (Yes route from S275), the conveyer-controlling module 2402 restarts the collector conveyor BC101 via the conveyor controller of the automatic baggage handling system Bap (step S280).
Note that an ‘Accept’ button may also be indicated on the touch screen 2305 at the step S250. In this configuration, if the Accept button is touched by an operator, the collector conveyor BC101 is forcibly restarted bypassing the step S255 and reaching the step S265. Nevertheless, executing this control shown as the Accept route from the step S250 is normally prohibited because the conveyance object Cob3 is resultantly conveyed in the downstream direction Ad without an operator inputting the overriding command.
‘Intrusion control’ based on an ‘intrusion image’ captured by the CCTV camera 2309 will be described.
In this intrusion control, the intrusion-analysing module 2407 of the secondary computer 2308 substantively always analyses the image stream (intrusion image) continuously supplied from the CCTV camera 2309 via the Ethernet switch 2302 (step S300 of
If the intrusion-analysing module 2407 detects the presence of a human within the region of interest (i.e. the ‘keep-out zone ZK’) within the view of the image stream captured by the CCTV camera 2309, the intrusion-analysing module 2407 immediately stops the collector conveyor BC101 via the conveyer-controlling module 2402 of the primary computer 2301 (step S310).
After that, if an operator confirms that there is no human in the keep-out zone ZK (Yes route from step S315), the intrusion-analysing module 2407 restarts operation of the collector conveyor BC101 via the conveyer-controlling module 2402 of the primary computer 2301.
In the example shown in
For example, if a human presence is detected in the region of interest within the view of the image stream captured by the CCTV camera 2309, but subsequently no human presence is detected in the region of interest, then the intrusion-analysing module 2407 may automatically restart operation of the collector conveyor BC101 via the conveyer-controlling module 2402 of the primary computer 2301.
The third embodiment of the present invention will now be described below mainly using
A conveyor supervisor 3000 according to the third embodiment has substantially the same configuration as the conveyor supervisor 2000 according to the second embodiment. Accordingly, differences between the conveyor supervisors 2000 and 3000 will be mainly described here and redundant description will be omitted.
As shown in
Specifically, as shown in
The right RGB camera 3206 is also a Global Shutter RGB type camera with a fixed manual focus lens
Due to such a hardware difference, in this embodiment, as shown in
The first upstream-assessment zone Z1u is an area in which an image of the conveyance object Cob (i.e. object image) is captured as the ‘first image’ by each of the left three-dimensional camera 2203 and the left RGB camera 2204 while the conveyance object Cob is conveyed by the collector conveyor BC101.
The second downstream-assessment zone Z2d is provided in the downstream direction Ad with respect to the first upstream-assessment zone Z1u as well as the conveyor supervisor 3000.
The second downstream-assessment zone Z2d is an area in which the object image is captured as the ‘second image’ by each of the right three-dimensional camera 3205 and the right RGB camera 3206 when the conveyance object Cob is stationary on the collector conveyor BC101 since the collector conveyor BC101 is stopped by the primary computer 2301.
The left light 2207 is a LED lighting unit that illuminates the first upstream-assessment zone Z1u so that the left three-dimensional camera 2203 and the left RGB camera 2204 can clearly capture the first image of the conveyance object Cob.
The right light 3208 is another LED lighting unit that illuminates the second downstream-assessment zone Z2d so that the right three-dimensional camera 3205 and the right RGB camera 3206 can clearly capture the second image of the conveyance object Cob.
As shown in
However, software modules of the primary computer 3301 are slightly different from each software module of the primary computer 2301 in the second embodiment.
Further, software modules of the AI vision processor 3311 are also slightly different from each software module of the AI vision processor 2311 in the second embodiment.
As shown in
An image-analysing module 3401, which is a software module designed for the AI vision processor 3311, is stored in a storage device (not shown in FIGs) of the AI vision processor 3311.
The conveyer-controlling module 3402 stops the collector conveyor BC101 such that the conveyance object Cob is positioned stationary within the second downstream-assessment zone Z2d if the image-analysing module 3401 of the AI vision processor 3311 determines based on the first image that the conveyance state of the conveyance object Cob does not satisfy the forwarding condition.
Further, the conveyer-controlling module 3402 restarts the collector conveyor BC101, which is stopped by the conveyer-controlling module 2402, if the image-analysing module 3401 determines based on the second image that the conveyance state of the conveyance object Cob satisfies the forwarding condition.
In the meantime, in the present embodiment, the determination on the conveyance state of the conveyance object Cob in the first upstream-assessment zone Z1u is referred to as the ‘primary determination’. On the other hand, the determination on the conveyance state of the conveyance object Cob in the second downstream-assessment zone Z2d is referred to as the ‘secondary determination’.
The overriding module 3403 executes an ‘overriding control’ that forcibly restarts the collector conveyor BC101 from the stationary state.
Specifically, this overriding operation is carried out under a situation where the image-analysing module 3401 of the AI vision processor 3311 determined that the conveyance state of the conveyance object Cob was not satisfying the forwarding condition, but an operator completed inputting the overriding command into the touch screen 2305.
The AI vision processor 3311, like the AI vision processor 2311, is a computer dedicated for neural net artificial intelligence (AI) processing.
The AI vision processor 3311 receives the data of the first images captured by the left three-dimensional camera 2203 and the left RGB camera 2204 from the primary computer 3301. Also, the AI vision processor 3311 receives the data of the second images captured by the right three-dimensional camera 3205 and the right RGB camera 3206 from the primary computer 3301.
The image-analysing module 3401 implements the learning model 3405 which has been preliminarily trained for the conveyor supervisor 3000.
The image-analysing module 3401 assesses and determines whether the conveyance state of the conveyance object Cob satisfies the forwarding condition based on the received image data and the trained learning model 3405.
The learning model 3405 defines an assessment standard. Specifically, the learning model 3405 is a machine learning model generated based on a vast amount of raw data (training data), in which a number of conveyance objects are photographed individually, and a learning data set corresponding to the factors of the conveyance objects.
The AI vision processor 3311 transmits results of the assessment and determination process to the primary computer 3301.
A second photo-eye sensor 3211 is provided beside an upstream edge of the second downstream-assessment zone Z2d.
The second photo-eye sensor 3211 is an optical sensor detecting that the conveyance object Cob has entered the second downstream-assessment zone Z2d.
The second photo-eye sensor 3211 is connected to the I/O controller 2304 with a digital I/O cable.
Further, the I/O controller 2304 is communicably connected to the primary computer 3301 via the Ethernet switch 2302.
Hence, the primary computer primary computer 3301 can immediately detect that the conveyance object Cob enters the second downstream-assessment zone Z2d by the second photo-eye sensor 3211.
With the above-mentioned arrangement, the conveyor controlling apparatus according to the third embodiment of the present invention provides the following advantages.
The analysis control of the conveyance object Cob by the conveyor supervisor 3000 is mainly performed in accordance with a main flowchart shown in
Note that, in parallel with the analysis control of the conveyance object Cob shown in
As shown in
Immediately subsequent to this, the conveyer-controlling module 3402 stays on standby to make a very short delay for ensuring that the entire conveyance object Cob enters the first upstream-assessment zone Z1u (step S310).
At this time, the conveyer-controlling module 3402 of the primary computer 3301 obtains the first image of the conveyance object Cob3 captured by each of the left three-dimensional camera 2203 and the left RGB camera 2204 (step 315). During this time, the collector conveyor BC101 is still running, hence, the conveyance object Cob3 keeps travelling in the downstream direction Ad.
Then, the image-analysing module 3401 of the AI vision processor 3311 determines whether the conveyance state of the conveyance object Cob3 satisfies the forwarding condition based on the first image captured by each of the left three-dimensional camera 2203 and the left RGB camera 2204 (steps S315 and S320 in
Also, at this time, the training module 3404 records the analysis results of the conveyance object Cob3 carried out by the image-analysing module 3401 of the AI vision processor 3311 and the captured first image of the conveyance object Cob3 in the external storage server (steps S315 and S320 in
If the conveyance state of the conveyance object Cob3 satisfies the forwarding condition (Yes route from step S320 in
On the other hand, if the conveyance state of the conveyance object Cob3 does not satisfy the forwarding condition (No route from step S320 in
The conveyer-controlling module 3402 recognises that the conveyance object Cob3 has entered the second downstream-assessment zone Z2d in response to the second photo-eye sensor 3211 detecting the conveyance object Cob3.
Nevertheless, the conveyer-controlling module 3402 makes a short delay to ensure that the entire conveyance object Cob3 enters the second downstream-assessment zone Z2d (step S325 in
After that, at step S335 of
If the operation mode of the conveyor supervisor 3000 with regard to restarting the collector conveyor BC101 is ‘automatic’ (Automatic route from step S345), the conveyer-controlling module 3402 obtains the second image of the conveyance object Cob3 captured by each of the right three-dimensional camera 3205 and the right RGB camera 3206 (step S370 in
Further, the image-analysing module 3401 of the AI vision processor 3311 determines whether the conveyance object Cob3 satisfies the forwarding condition based on the second image captured by the right three-dimensional camera 3205 and the right RGB camera 3206 (steps S370 and S375 in
Also, at this time, the training module 3404 records the analysis results of the conveyance object Cob3 carried out by the image-analysing module 3401 of the AI vision processor 3311, the determination results as to whether or not the conveyance state of the conveyance object Cob3 satisfies the forwarding condition, and the captured second image of the conveyance object Cob3 in the external storage server (steps S370 and S375 of
If the conveyance state of the conveyance object Cob3 satisfies the forwarding condition (Yes route from S375), the conveyer-controlling module 3402 restarts the collector conveyor BC101 via the conveyor controller of the automatic baggage handling system Bap (step S380). At this time, in the example shown in
On the other hand, if the operation mode of the conveyor supervisor 3000 with regard to restarting the collector conveyor BC101 is ‘manual’ (Manual route from step S345), the conveyer-controlling module 3402 indicates a ‘Force-in’ button and a ‘Retry’ button on the touch screen 2305.
If the Force-in button is touched by an operator (see Force-in route from step S350 in
In other words, this situation could be, for example, that although an operator has checked the conveyance object Cob3 at step S340, no problem was actually found, and therefore the operator believes that the conveyance object Cob3 should continue travelling in the downstream direction Ad.
At this time, the operator appropriately checks the check boxes corresponding to the reasons why there is no problem with the conveyance object Cob3 (i.e. ‘overriding reason’), then the operator touches a ‘Submit’ button indicated on the touch screen 2305 (step S355).
Also, at this time, the training module 3404 records the analysis results of the conveyance object Cob3 carried out by the image-analysing module 3401 of the AI vision processor 3311, the determination results as to whether or not the conveyance state of the conveyance object Cob3 satisfies the forwarding condition, the captured first image of the conveyance object Cob3, and the overriding reason entered through the touch screen 2305 in the external storage server (step S360).
In addition, the training module 3404 may apply a learning control to the learning model 3405 based on the entered overriding reason.
In short, in the situation where the image-analysing module 3401 of the AI vision processor 3311 has determined that the conveyance state of the conveyance object Cob3 does not satisfy the forwarding condition, but an operator has completed inputting the overriding command through the touch screen 3305, accordingly, the overriding module 3403 executes the overriding control to forcibly restart the collector conveyor BC101 via the conveyor controller of the automatic baggage handling system Bap (step S365 of
In contrast, at the step S350, if an operator touches the Retry button (see Retry route from step S350), the conveyer-controlling module 3402 obtains the second image of the conveyance object Cob3 captured by each of the right three-dimensional camera 3205 and the right RGB camera 3206.
In other words, this situation could be, for example, that although an operator has checked the conveyance object Cob3 in the second downstream-assessment zone Z2d at the step S340, no problem was actually found, and therefore he/she wants to have the conveyor supervisor 3000 reconfirm the conveyance state of the conveyance object Cob, as a precaution.
As another example situation, an operator has solved the problems of the conveyance object Cob3 in the second downstream-assessment zone Z2d at the step S340, and as a precaution, the operator wants to make sure that the correction to the conveyance object Cob3 was appropriate by using the conveyor supervisor 3000.
If the Retry button is touched (see Retry route from step S350), the image-analysing module 3401 of the AI vision processor 3311 determines, based on the second image captured by each of the right three-dimensional camera 3205 and the right RGB camera 3206, whether or not the conveyance state of the conveyance object Cob3 satisfies the forwarding condition (steps S370 and S375).
If the conveyance state of the conveyance object Cob3 satisfies the forwarding condition (Yes route from S375), the conveyer-controlling module 3402 restarts the collector conveyor BC101 via the conveyor controller of the automatic baggage handling system Bap (step S380).
Note that an ‘Accept’ button may also be indicated on the touch screen 2305 at the step S350. In this configuration, if the Accept button is touched by an operator, the collector conveyor BC101 is forcibly restarted bypassing the step S355 and reaching the step S365. Nevertheless, executing this control shown as the Accept route from the step S350 is normally prohibited because the conveyance object Cob3 is resultantly conveyed in the downstream direction Ad without an operator inputting the overriding command.
Regarding the intrusion control using the CCTV camera 2309 has been described in the second embodiment, so description thereof will be omitted here.
The present invention is not limited to the above-described embodiments and the variants thereof, and can be variously modified and implemented.
In the first to third embodiments, the conveyor supervisors 1000/2000/3000 are installed in the airport AP, but is not limited to this scenario. For example, the conveyor supervisors 1000/2000/3000 may be installed in factories, plants, and/or warehouses.
In the first to third embodiments, the conveyor supervisors 1000/2000/3000 are installed near the collector conveyor BC101, partly because, normally, there are more ground-handling staff working in the FOH than the BOH. Accordingly, if the conveyor controlling apparatus Baz is installed in the FOH of the airport AP, then the conveyance object Cob may be quickly attended by the ground-handling staff in the FOH.
Nevertheless, the installation point of the conveyor supervisors 1000/2000/3000 is not limited to this location. For example, the conveyor supervisors 1000/2000/3000 may be installed near the transport conveyor BC102 close to the inlet of the BOH in the airport AP.
In the first and third embodiments, as shown in
That is, as shown in
In this case, the first upstream-assessment zone Z1u on the collector conveyor BC101 is a virtual area that approximately corresponds to a photographable area of the cameras towards the upstream direction Au (i.e. each of the right three-dimensional cameras 205/3205 and each of the right RGB cameras 206/3206). Likewise, the second downstream-assessment zone Z2d on the collector conveyor BC101 is another virtual area that approximately corresponds to a photographable area of the cameras towards the downstream direction Ad (i.e. each of the left three-dimensional cameras 203/2203 and each of the left RGB cameras 204/2204).
The conveyor supervisors 1000/2000/3000 are robust and can work with belt conveyors, roller conveyors, slat conveyors, mesh conveyors and the like. The conveyor supervisors 1000/2000/3000 can also be adapted to work in scenarios where the conveyor has been replaced by robot-type vehicles such as Automated Guided Vehicles (AGVs). Therefore, again, the conveyor supervisors 1000/2000/3000 are not limited for airports. Applicable industries are not limited to the airline sector, and can be applied to various industries.
As a variant of the first to third embodiments, the conveyor supervisors 1000/2000/3000 may recognise a baggage ID printed on a baggage tag which is uniquely affixed to the conveyance object Cob, based on the first image and/or the second image.
In this case, the baggage ID recognised by the conveyor supervisors 1000/2000/3000 may be transmitted to other external systems (e.g. systems under the control of the International Air Transport Association (IATA), other cargo systems, and so on).
Further, the first and/or the second images may be transmitted to other external systems along with the baggage ID recognised by the conveyor supervisors 1000/2000/3000.
Further, the conveyor supervisors 1000/2000/3000 may transmit the first image and/or the second image of the conveyance object Cob, which did not satisfy the forwarding condition, to other external systems.
In the first to third embodiments, the LED illumination of the status indicators 306/2306 is changed in accordance with the working state of the conveyor supervisors 1000/2000/3000, but it is not limited to this configuration. For example, the conveyor supervisors 1000/2000/3000 may include a buzzer (not shown in FIGs) as an alarm device for beeping in accordance with the working state of the conveyor supervisors 1000/2000/3000. According to this configuration, it is possible to notify aurally an operator of the operational state of the conveyor.
The conveyor supervisors 1000/2000/3000 may be movable. For example,
The primary computers 301/2301/3301 may determine the conveying direction Ac of the collector conveyor BC101 based on the first image and/or the second image. That is, the conveyor supervisors 1000/2000/3000 may judge the conveying direction Ac of the collector conveyor BC101 is either rightwards (see
In the first and third embodiments, respective first photo-eye sensors 209/2209 are provided beside the first upstream-assessment zone Z1u, and respective second photo-eye sensors 211/3211 are provided beside the second downstream-assessment zone Z2u. Further, in the second embodiment, the first photo-eye sensor 2209 is provided is beside the first upstream-assessment zone Z1u, and the second photo-eye sensor 2211 is provided is beside the second upstream-assessment zone Z2u.
However, it is not limited to theses configurations.
For example, without using the first photo-eye sensors 209/2209, the primary computers 301/2301/3301 according to the first to third can detect an entry of a conveyance object Cob into the first upstream-assessment zone Z1u based on the first image.
Similarly, even without using the second photo-eye sensors 211/3211, the primary computers 301/3301 according to the first and third embodiments can detect an entry of a conveyance object Cob into the second upstream-assessment zone Z2u based on the downstream image.
Likewise, without using the second photo-eye sensor 2211, the primary computer 2301 according to the second embodiment can detect an entry of a conveyance object Cob into the second downstream-assessment zone Z2d based on the second image.
Nevertheless, by using the first and second photo-eye sensors 209, 221, 2209, 2211 and 3211, the processing load of the conveyor supervisors 1000/2000/3000 may be reduced.
In the first to third embodiments, the height of a conveyance object is detected based on the captured image by the three-dimensional camera. However, it is not limited to this configuration.
For example, an optical height sensor may be provided beside the conveyor.
According to this configuration, it is possible to judge more directly whether or not the height of the conveyance object Cob is beyond the predetermined height limit.
In the second embodiment, the image-analysing module 2401 of the AI vision processor 2311 analyses and determines whether the conveyance state of the conveyance object Cob satisfies the forwarding condition.
Likewise, in the third embodiment, the image-analysing module 3401 of the AI vision processor 3311 analyses and determines whether the conveyance state of the conveyance object Cob satisfies the forwarding condition.
However, it is not limited to these configurations.
For example, the technology described in the second embodiment may be realised by a single computer that integrates all the features of the primary computer 2301 and the AI vision processor 2311.
Also, the technology described in the third embodiment may be realised by a single computer that integrates all the features of the primary computer 3301 and the AI vision processor 3311.
In the first to third embodiments, the secondary computers 308/2308 carry out the intrusion control using the CCTV camera 212/2309. However, it is not limited to these configurations.
For example, the technology described in the first to third embodiments may be realised by a single computer that integrates all the features of the primary computers 301/2301/3301 and the secondary computers 308/2308.
In the first to third embodiments, the case where the conveyor supervisors 1000/2000/3000 are applied to the straight collector conveyor BC101 has been described, but it is not limited to this. For example, the conveyor supervisors 1000/2000/3000 are applicable to curbed conveyors.
The situations in which a conveyance state of a conveyance object Cob does not satisfy the forwarding conditions as discussed in the first to third embodiments may corresponds to at least one of ‘rejection categories’ shown in
In the first to third embodiments, the data related to analysis/determination of the conveyance object Cob3, the data related to the intrusion control, and the data related to the captured images of the conveyance object Cob3 is recorded to the external storage server, but it is not limited to this configuration.
For example, it is also possible to employ a configuration in which the data relating to the analysis/determination of the conveyance object Cob, the intrusion control, and the data related to the captured images of the conveyance object Cob is recorded to an internal storage device (now shown in FIGs) provided in the conveyor supervisor 1000/2000/3000.
As described in detail above, according to the control device and control method of the present invention, the conveyor is controlled according to the environmental parameters of the conveyor, so that the reliability of the conveyance system having the conveyor can be improved.
In addition, by stopping the conveyor according to different forwarding conditions depending on the conveyance system having the conveyor and the image of the conveyance objects, it is possible to improve the accuracy of flow control for the conveyance objects. Further, it is possible to improve overall operational efficiencies of the conveyance system.
Further, when it is later determined that the conveyance state of the conveyance object now satisfies the forwarding condition of the conveyor, the conveyer can be quickly restarted.
Also, the conveyance state of the conveyance object may be determined in the first assessment zone, and the conveyor may be stopped so that the conveyance object is positioned within the second assessment zone according to the determination result. Further, the conveyance object within the second basement zone may be further determined, and the conveyor may be restarted depending on the determination result.
Note that the first assessment zone may be defined on the upstream side of the capturing device, and the second assessment zone may be defined on the downstream side of the capturing device.
Also, the learning model may be used in determining the image of the conveyance object.
Moreover, even if it is determined that the conveyance state of the conveyance object does not satisfy the forwarding condition, the control for forcibly restarting the conveyor, that is, the overriding control may be executed. However, when executing this overriding control, the reason for the execution must be included, so that more accurate control of the conveyance object can be realised.
Further, it improves the operational reliability of the transport system having the conveyor by immediately stopping the conveyor if any human is detected with in the keep-out zone.
Also, the conveyor can be restarted quickly if no humans are detected within the keep-out zone.
In addition, by making the conveyor controlling apparatus compact with the main body, the apparatus can be adapted to various types of conveyors.
In the claims which follow and in the preceding description of the invention, except where the context requires otherwise due to express language or necessary implication, the word “comprise” or variations such as “comprises” or “comprising” are used in an inclusive sense, i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.
The reference to any prior art in this specification is not, and should not be taken as, an acknowledgement or any form of suggestion that the prior art forms part of the common general knowledge.
| Number | Date | Country | Kind |
|---|---|---|---|
| PCT/AU2021/051420 | Nov 2021 | WO | international |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/AU2022/051399 | 11/23/2022 | WO |
