Patent Application
20220144609
This application is based upon and claims the benefit of priority from Japanese Patent Application No. 2020-186088, filed Nov. 6, 2020, the entire contents of which are incorporated herein by reference.
Embodiments described herein relate generally to an autonomous mobile robot, a transporter, an autonomous mobile robot control method, and a transporter control method.
Conventionally, an AGV (Automatic Guided Vehicle) has been used. For example, the AGV is configured to calculate a self-location by measuring a direction and a distance with respect to a reflector provided on a wall using a laser range finder, and automatically travels in accordance with the calculation result.
Furthermore, a technique of simultaneously carrying out estimation of localization and formation of an environment map, which is referred to as SLAM (Simultaneous Localization and Mapping), is known.
According to one embodiment, an autonomous mobile robot has a driver, a first detector, a second detector, a localization estimation part, a route-generating part, and a control part. The autonomous mobile robot is autonomously movable. The driver is configured to cause the autonomous mobile robot to move. The first detector is provided at a position higher than a position in height of an object present around the autonomous mobile robot. The first detector is configured to obtain first data by scanning the object at a first region around the autonomous mobile robot. The second detector is configured to obtain second data by scanning the object at a second region around the autonomous mobile robot. The localization estimation part is configured to calculate an estimated position of the autonomous mobile robot in a predetermined region in accordance with the first data. The route-generating part is configured to calculate a position of the object present around the autonomous mobile robot in accordance with the second data. The route-generating part is configured to calculate a route to a target position in accordance with the estimated position and the position of the object. The control part is configured to control the driver in accordance with the route. The control part is configured to cause the autonomous mobile robot to travel to the target position.
Hereinafter, an autonomous mobile robot, a transporter, an autonomous mobile robot control method, and a transporter control method according to an embodiment will be described with reference to the drawings. In the embodiment, the autonomous mobile robot is applicable to a transporter. In the embodiment, the transporter is configured to transport a transport object. The transporter transfers the transport object in which packages, commercial products, or the like are loaded in a distribution center or a back yard of a shop.
As shown in
A transport object (not shown in the drawings) is to be mounted on the upper surface side of the bottom plate D1. The frame D2 is provided so as to surround the transport object on the upper surface side of the bottom plate D1. The frame D2 has four faces. The frame D2 is formed so as to surround a circumference of the transport object, for example, on four surfaces of the transport object above the bottom plate D1. At least one of the four faces of the frame D2 is an openable and closable door. The frame D2 may have a configuration in which one face of the frame D2 is opened without providing a door.
Four wheels C are disposed at four corners on the lower surface side of the bottom plate D1. The wheels C cause the bottom plate D1 to move with respect to the floor surface E. The wheels C are two pairs of wheels provided on the lower surface side of the bottom plate D1. The two pairs of the wheels are spaced at a predetermined distance in the travel direction P. With the aforementioned configuration, the transport object D travels on the floor surface E when an external force is applied thereto in the travel direction P. The transport object D is transferred to a target position by the transporter 1.
The transporter 1 includes, for example, a transfer carriage 2 that transfers the transport object D and a plurality of detectors used for sensing provided at the transfer carriage 2. The detectors includes a first detector 10, a second detector 11, and a third detector 12. The transfer carriage 2 includes, for example, a load base 3 having an upper surface. The bottom plate D1 is to be mounted on the upper surface of the load base 3. The load base 3 has a lower surface. A plurality of wheels 6 are provided on the lower surface of load base 3. The wheels 6 are driven by a driver 14 (refer to
The wheels 6 are driven by a commonly-used two-wheel independent drive system using the driver 14 described below. As the wheels 6 are driven, the transporter 1 travels. The direction of movement of the transporter 1 is controlled by a control part 110 as described hereinbelow. The wheels 6 may be driven by use of a steering wheel or a particular kind of wheel so as to omnidirectionally travel.
The height of the load base 3 is determined such that the load base 3 can enter into a region between the bottom plate D1 of the transport object D and the floor surface E. A lifter 4 is provided on the upper surface side of the load base 3. The height of the lifter 4 is adjustable. The lifter 4 is controlled by the control part 110 described later.
When the load base 3 is inserted between the bottom plate D1 and the floor surface E, the lifter 4 increases the position thereof in height and lifts up the transport object D. The load base 3 transfers the transport object D in a state in which the transport object D is lifted up by the lifter 4. The load base 3 may transfer the transport object D while generating a friction between the lower surface side of the bottom plate D1 and an upper surface side of the lifter 4 without lifting up the transport object D. The transfer carriage 2 may transfer the transport object D by traction while hooking the transport object D using a pin or the like without providing the load base 3 and the lifter 4.
A housing 7 is provided at, for example, a front side in the travel direction P on the load base 3 (refer to
A support member 8 is provided at an upper surface side of the housing 7. The support member 8 is formed in a rod shape. The support member 8 is configured to be provided upright. The support member 8 is formed in, for example, a frame shape when viewed from a direction along the travel direction P. Particularly, the support member 8 has a configuration in which two rod members are provided upright, and another rod member is further connected to the two rod members so as to intersect therewith. The support member 8 may be formed so as to have a reinforced truss structure that increases bending rigidity in a front-back direction for preventing vibration or may be formed so as to have other structures.
The first detector 10 and the rotating light K are provided at an upper end of the support member 8. The first detector 10 detects an object around the transporter 1. The first detector 10 is, for example, an LRF (Laser Range Finder). The first detector 10 scans the object with a laser beam, receives reflection light reflected from the object, and measures a distance to the surface of the object in accordance with a phase difference of the reflection light or a differential arrival time. The first detector 10 obtains first data by scanning the object at a first region around the transfer carriage 2. The first detector 10 scans the object in a predetermined angle range, for example, in a planar direction around the first detector 10 and obtains the first data associated with a distance to the surface of the object around the first detector 10 at a plurality of points.
The first data is used to generate a localization-and-estimation map of the transfer carriage 2 as described hereinbelow. In the case in which there is a difference in an ambient environment between a time of generating the localization-and-estimation map and a time of practically carrying out a transfer operation, a degree of accuracy of estimating the localization becomes degraded. In a distribution center, a back yard of a shop, or the like, a human is present around the transporter 1, other roll box pallets are placed, and the ambient environment of the transporter 1 varies from hour to hour. Moreover, for example, in a distribution center or a back yard of a shop, commercial products are often on a store shelf or the like near the floor surface, and the number of products temporally varies.
Consequently, depending on the above, a situation of the object to be detected by the LRF may vary temporally. In the distribution center, for example, roll box pallets or operators frequently move, and the object to be detected by the LRF varies. The ambient environment includes an environment inherence object and a surrounding existence object. The environment inherence object means an inherence object, for example, a circumference wall, fixed equipment, a pillar, or the like. It is assumed that the position or the shape of the environment inherence object is not changed when the transfer carriage 2 is driven. The surrounding existence object is an object other than the environment inherence object of the ambient environment. It is assumed that the position or the shape of the surrounding existence object is changed when the transfer carriage 2 is driven. The surrounding existence object is, for example, an object present around the transfer carriage 2 (a human, a transport object such as transfer carriage, roll box pallet, or the like).
As compared with the environment inherence object, the surrounding existence object is often located closer to the moving transfer carriage 2 than the environment inherence object. Furthermore, regarding the ambient environment, the environment inherence object such as the circumference wall, the fixed equipment, the pillar, or the like is not changed. Consequently, the first detector 10 is attached to a position in height which is assumed in advance such that the surrounding existence object of the ambient environment such as a transfer object, a movable object, or the like is not present. Particularly, the first detector 10 is provided at a position in height of the object (surrounding existence object) present around the transfer carriage 2.
In an environment in which there are a lot of movable objects or operators are present, it is preferable that the first detector 10 be attached to a position higher than that of the other transport object (roll box pallet) or the moving obstacle such as a human, which causes the ambient environment to be varied. The first detector 10 is provided at a position, for example, higher than or equal to 1.8 meters, which is a position higher than a body height of a common person. Heights of roll box pallets used in a distribution center or a back yard of a shop are different from each other. The first detector 10 is provided such that the height of the first detector 10 is manually adjustable so as to be higher than the position in height of an object, for example, a moving obstacle or the like present around the first detector. Additionally, an adjuster (not shown in the drawings) may be provided at the first detector 10. The adjuster can automatically adjust the position of the first detector 10 to be higher than the position in height of an object, for example, a moving obstacle or the like present around the first detector.
The second detector 11 that obtains second data by scanning the object is provided at a second region around the transfer carriage 2 at the forward side of the housing 7 in the travel direction P. The second detector 11 is, for example, an LRF. The second detector 11 scans the object with a laser beam, receives reflection light reflected from the object, and measures a distance to the surface of the object in accordance with a phase difference of the reflection light or a differential arrival time. The second detector 11 scans the object in a predetermined angle range, for example, in a planar direction in front of the second detector 11 and obtains the second data associated with a distance to the surface of the object in front of the second detector 11 at a plurality of points. The second detector 11 detects, for example, a transfer object or a movable object around the second detector 11. The movable object means, for example, a shelf that is optionally changeable in position, the other transport object D that is being temporarily halted, a transfer object such as a human, or the like.
The third detector 12 is provided at the back side of the load base 3 in the travel direction P. The third detector 12 obtains third data associated with positions of the wheels 6 (leg) provided at the lower portion of the transport object D. The third detector 12 is, for example, an LRF. The third detector 12 is attached to a lower position in order to detect the wheels 6. The third detector 12 scans the wheels 6 with a laser beam, receives reflection light reflected from the wheels 6, and measures a distance to the surface of the wheel 6 in accordance with a phase difference of the reflection light or a differential arrival time.
The third detector 12 scans the object in a predetermined angle range, for example, in a planar direction of the laser scanning direction and obtains the third data associated with a distance to the surface of the wheel 6 in front of the third detector 12 at a plurality of points. As described below, the positions of the wheels 6 supporting the transport object D are obtained in accordance with the third data detected by the third detector 12, and the load base 3 can be inserted between the pair of the wheels 6 facing each other in the travel direction P. A depth camera capable of obtaining distance information as well as the LRF may be used as the third detector 12. The third detector 12 may be configured of a plurality of infrared distance sensors.
Next, control of the transporter 1 will be described.
As shown in
The controller 100 includes, for example, a storage 102, a map-generating part 104, a localization estimation part 106, a route-generating part 108, a control part 110, and a docking control part 112. The storage 102 stores various data. The map-generating part 104 generates a map for estimating a position of the transporter 1. The localization estimation part 106 calculates an estimated position of the transfer carriage 2 in accordance with the first data obtained by the first detector 10 and the map. The route-generating part 108 calculates a route to a target position of the transporter 1 in accordance with the second data obtained by the second detector 11 and the estimated position. The control part 110 controls the driver 14 in accordance with the third data formed by the route and the third detector 12 and causes the transfer carriage to travel to the target position. The docking control part 112 controls docking with respect to the transport object D.
For example, as a processor such as CPU (Central Processing Unit) executes program (computer program, software) stored in the storage 102, a part or all of functional parts of the map-generating part 104, the localization estimation part 106, the route-generating part 108, the control part 110, and the docking control part 112 is operated. Furthermore, a part or all of the functions of the aforementioned constituent parts may be achieved by hardware (circuit part, including circuitry) such as LSI (Large Scale Integration Circuit), ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Gate Array), GPU (Graphics Processing Unit), or the like, or may be achieved by cooperation of software and hardware. The program may be stored in advance in a storage device such as a HDD (hard disk drive), a flash memory, or the like. The program may be stored in a removable recording medium such as a DVD, a CD-ROM, or the like. By loading the recording medium to a drive device, the program may be installed in the drive device.
A localization-and-estimation map is generated by SLAM in order to estimate the localization of the transporter 1. In the SLAM, it is assumed that the environment of generating the map is the same as the environment of movement of the locomotion robot. Initially, the localization-and-estimation map is generated by causing the transporter 1 to travel by a manual operation. When the transporter 1 travels in a predetermined region such as a business place, that is an object in which the transporter 1 travels, the first data generated by the first detector 10 is stored in the storage 102. The storage 102 is realized by a recording medium, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), a HDD, a flash memory, or the like.
The map-generating part 104 uses, for example, the first data, calculates a two-dimensional shape associated with the environment inherence object such as the circumference wall, the fixed equipment, or the like, and generates the localization-and-estimation map. The map-generating part 104 may generate the localization-and-estimation map in accordance with the first data as well as the localization-and-estimation map in accordance with the second data. The map-generating part 104 may separately generate the localization-and-estimation map at the horizontal plane at a measured height of the first detector 10 in accordance with construction drawing data. The localization-and-estimation map generated by the map-generating part 104 is stored in the storage 102.
The localization estimation part 106 calculates two-dimensional distance information associated with the environment inherence object such as the wall around the transporter 1 (the transfer carriage 2), the fixed equipment, or the like in accordance with the first data obtained by the first detector 10. The localization estimation part 106 compares the calculated two-dimensional distance information to the localization-and-estimation map and extracts the predetermined region coincident with the two-dimensional distance information from the localization-and-estimation map.
The localization estimation part 106 calculates a relative angle (posture) and a relative position with respect to a planar direction of the transporter 1 in the predetermined region extracted in accordance with the extracted predetermined region and the calculated two-dimensional distance information. The localization estimation part 106 calculates an estimated position on the two-dimensional coordinate system of the transporter 1 in the localization-and-estimation map in accordance with the calculation result.
The route-generating part 108 calculates the shape of the obstacle present around the transporter 1 in accordance with the second data obtained by the second detector 11. The route-generating part 108 compares the two-dimensional distance information associated with the environment inherence object such as the wall around the transporter 1, the fixed equipment, or the like in the extracted predetermined region and the shape of the object present around the transporter 1 in accordance with the second data, and extracts the obstacle present around the transporter 1. The route-generating part 108 generates a route from the estimated position of the transporter 1 to the target position.
At this time, the route-generating part 108 generates a route that avoids the extracted obstacle. In accordance with the second data, the route-generating part 108 updates real-time information associated with the obstacles such as the number, the position, the direction of movement, the speed, the size, or the like at a predetermined timing, and updates route that avoids the obstacle. The predetermined timing is adjusted in accordance with the movement speed of the transporter 1. The route-generating part 108 may not only generate a route that avoids the obstacle in accordance with a moving state of the obstacle but also generate a route that avoids the obstacle by adjusting the speed of the transfer carriage 2
The control part 110 controls the driver 14 in accordance with the route generated by the route-generating part 108 and causes the transfer carriage 2 to travel along the route. The driver 14 independently controls the wheels 6 provided at right and left of the transfer carriage 2, and controls the speed and the direction of the transfer carriage 2. The control part 110 controls the driver 14 and thereby controls the transfer carriage 2 so as to be stopped or so as to avoid the obstacle in the case in which the obstacle comes close at a distance to be less than or equal to a previously-set reference. At this time, in the case in which the transport object D being transferred, the control part 110 controls deceleration of the driver 14 and the direction of the transfer carriage so as to prevent load collapse.
The control part 110 may cause the rotating light K to change a light emission pattern thereof and to light up in the case in which a human comes close at a distance to be less than or equal to a previously-set reference. As well as the rotating light K, the control part 110 may cause a speaker (not shown in the drawings) to emit sound and thereby notify the human. In the case in which the obstacle is the other transporter 1, the control part 110 may communicate with the other transporter 1 and adjust the relative distance with respect to the other transporter 1. The control part 110 may communicate with a control apparatus (not shown in the drawings) via a network and adjust the relative position with respect to the other transporter 1.
The docking control part 112 recognizes arrangement of the wheels C of the transport object D in accordance with the third data obtained by the third detector. The docking control part 112 recognizes a space between the wheels C on the lower surface side of the transport object D in accordance with the third data. The docking control part 112 recognizes the relative position and the posture of the transport object D. The distance between the pair of the wheels C facing each other in the travel direction P of the transport object D is different from distance between the pair of the wheels C facing each other in the direction orthogonal to the travel direction P. The docking control part 112 determines the insertion direction of the transfer carriage 2 with respect to the transport object D in accordance with the distance between the pair of the wheels C in the recognized travel direction P and the distance between the pair of the wheels C in the direction orthogonal to the travel direction P. The docking control part 112 generates a docking route in which the transfer carriage 2 is inserted between the wheels C in accordance with the third data.
The control part 110 controls the driver 14 in accordance with the docking route generated by the docking control part 112 and causes the transfer carriage 2 to be disposed at the transport object D. The control part 110 controls the driver 14 and causes the transfer carriage 2 to be inserted into the lower surface side of the transport object D from the back portion thereof. When the transfer carriage 2 reaches a predetermined position at the lower surface side of the transport object D, the control part 110 causes the driver 14 to be stopped, controls the lifter 4 so as to lift up the bottom plate D1 of the transport object D, and causes the transport object D to be in a state of being transportable.
Next, a controlling method of the transporter 1 will be described.
The docking control part 112 recognizes a space between the wheels C of the transport object D in accordance with third data obtained by the third detector 12. The control part 110 causes the transfer carriage 2 to be disposed in the space at the lower side of the transport object D in accordance with the recognized space (step S102). The localization estimation part 106 calculates a shape of a peripheral region in accordance with the first data obtained by the first detector 10, compares the calculation result and the localization-and-estimation map, and extracts a predetermined region coincident with the localization-and-estimation map.
The localization estimation part 106 calculates an estimated position of the transfer carriage in the extracted predetermined region (step S104). The route-generating part 108 calculates a shape and a position of the peripheral region in accordance with the second data obtained by scanning the object at a second region around the transfer carriage 2, and extracts an obstacle around the transfer carriage 2 in accordance with the calculate shape and the localization-and-estimation map (step S106). The route-generating part 108 generates a route to a target position in accordance with the estimated position and generates a route to the target position, which avoids the obstacle in accordance with the first data and the second data (step S108). The control part 110 controls the driver 14 in accordance with the route and causes the transfer carriage 2 to travel to the target position (step S110). In the case in which the other transport object is present, the steps after the step 102 are repeated.
As described above, according to the transporter 1, in a distribution center or a back yard of a shop, it is possible to automatically transfer the transport object D such as a roll box pallet or the like, in which, packages, commercial products, or the like are loaded. According to the transporter 1, since the first detector 10 is provided at the position higher than that of the obstacle or the like around the first detector, it is possible to calculate the estimated position of the transporter 1 without being affected by the obstacle. Since the second detector 11 detects the obstacle therearound, the transporter 1 can travel along the route to the target position while avoiding the obstacle.
Hereinbelow, a second embodiment will be described. In the following description, identical names and identical reference numerals are used for the components which are identical to those of the above-described embodiment, and duplicate description is omitted (the same applies hereinafter).
As shown in
The controller 100 may include a recognition part 114 that recognizes the transport object D and an object present around the transfer carriage 2 in accordance with an image data obtained by the camera R. The recognition part 114 extracts an obstacle around the transfer carriage 2 in accordance with the image data. The recognition part 114 may recognize the transport object D in accordance with the image data. The recognition part 114 extracts the transport object D in accordance with the image data, and recognizes a distance to the transport object D, a relative posture of the transport object D, and an insertion direction of the transfer carriage 2.
The control part 110 controls the driver 14 based on the recognition result recognized by the recognition part 114, causes the transfer carriage 2 to travel so as to be inserted into the lower surface side of the transport object D, and causes the transfer carriage 2 to be disposed at the lower surface side of the transport object D.
The route-generating part 108 generates a route to a target position, which avoids the obstacle in accordance with the position, the speed, the direction of movement, the size, or the like of the obstacle extracted by the recognition part 114. The control part 110 causes the transfer carriage 2 to travel to the target position while avoiding the obstacle in accordance with the generated route. The route-generating part 108 may extract the obstacle present in the circumference by use of the image data obtained by the camera R and the second data of the second detector 11. As the aforementioned camera R, a plurality of infrared distance sensors that measure a distance may be used.
As described above, according to the transporter 1A, by use of the camera R, it is possible to easily avoid the obstacle in the circumference. According to the transporter 1A, by use of the camera R, it is possible to dispose the transfer carriage 2 at the lower surface side of the transport object D. Additionally, when the transport object is unloaded at a transferring place, it is possible to arrange the transport object next to the other transport objects side by side.
In the aforementioned embodiment, an object present around the transfer carriage 2 is extracted by use of the second detector 11 or the camera R. In the third embodiment, the object around the transfer carriage 2 may be extracted by use of the first data of the first detector 10. In the case in which, for example, a three-dimensional LRF that can also carries out scanning in a vertical direction is used as the first detector 10, the route-generating part 108 may extract an object such as the obstacle or the like around the transfer carriage 2 in accordance with the first data obtained by the first detector 10.
As described above, according to the third embodiment, it is possible to extract the object around the transfer carriage 2 in accordance with the first data obtained by the first detector 10, and it is possible to simplify a device configuration.
The first detector 10 is attached to a high position via the support member 8. Consequently, there is a concern that the value of the first data varies due to an error in position at which the first detector 10 is attached or an effect of vibration generated along with movement of the transfer carriage 2. In the fourth embodiment, the controller 100 compares the second data obtained by the second detector 11 and the first data obtained by the first detector 10, and carries out calibration with respect to an error in position at which the first detector is attached or an error generated due to vibration.
At this time, a measurement object to be measured by the second detector 11 is selected from the environment inherence objects such as a wall, a pillar, or the like of a building, in which it is assumed that the shape at the detection position of the second detector 11 is the substantially same as that of the detection position of the first detector 10. Since the second detector 11 is provided at the housing 7 of the transfer carriage 2, the second data is stabilized more than the first data. For this reason, in the controller 100, it is possible to carry out calibration of the first data by carrying out verification by comparing the first data and the second data. The route-generating part 108 calculates the shape of the object of the circumference in accordance with the first data. The route-generating part 108 calculates the shape of the object of the circumference in accordance with the second data. The route-generating part 108 extracts characteristic parts, for example, the wall, the pillar, the corner, or the like of the building in accordance with the first data and the second data.
The route-generating part 108 compares a second characteristic portion of the building in accordance with the second data and a first characteristic portion of the building in accordance with the first data. The route-generating part 108 compares, for example, data associated with a distance, a posture, a shape, or the like of the second characteristic portion and data associated with a distance, a posture, a shape, or the like of the first characteristic portion, and thereby carries out calibration with respect to the error in position at which the first detector is attached or the error generated due to vibration.
According to the fourth embodiment, in the case of measuring the same measurement object by the second detector 11 and the first detector 10, the first data is compared to the second data stabilized more than the first data. As a result, it is possible to carry out the calibration with respect to the error in position at which the first detector 10 is attached or the error generated due to vibration which is included in the first data.
In the fourth embodiment, the calibration with respect to the error in position at which the first detector 10 is attached or the error generated due to vibration which is included in the first data is carried out by comparing the first data of the first detector 10 and the second data of the second detector 11. In the fifth embodiment, the first data is corrected by detecting swing of the first detector 10.
As shown in
In each of the aforementioned embodiments, although the map-generating part 104, the localization estimation part 106, the route-generating part 108, the docking control part 112, and the recognition part 114 are each software functional part, each of them may be a hardware functional part such as LSI or the like.
According to at least one embodiment described above, the autonomous mobile robot is autonomously movable, a driver 14 configured to cause the autonomous mobile robot to move; a first detector 10 provided at a position higher than a position in height of an object present around the autonomous mobile robot, the first detector being configured to obtain first data by scanning the object at a first region around the autonomous mobile robot; a second detector 11 configured to obtain second data by scanning the object at a second region around the autonomous mobile robot; a localization estimation part 106 configured to calculate an estimated position of the autonomous mobile robot in a predetermined region in accordance with the first data; a route-generating part 108 configured to calculate a position of the object present around the autonomous mobile robot in accordance with the second data, the route-generating part being configured to calculate a route to a target position in accordance with the estimated position and the position of the object; and a control part 110 configured to control the driver in accordance with the route, the control part being configured to cause the autonomous mobile robot to travel to the target position. Accordingly, the autonomous mobile robot can automatically travel in the environment in which a movable object is present around the autonomous mobile robot.
While certain embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. Indeed, the novel embodiments described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the embodiments described herein may be made without departing from the inventions. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope of the inventions.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2020-186088 | Nov 2020 | JP | national |
