Robotic cleaning device and a method at the robotic cleaning device of performing cliff detection

Abstract

A robotic cleaning device and a method at the robotic cleaning device of performing cliff detection along a surface over which the robotic cleaning device moves. The method includes illuminating the surface with at least one light source, capturing an image of the surface, detecting at least one illuminated section in the captured image, and determining distance to objects in the at least one illuminated section of the captured image. The method further comprises comparing at least two of the determined distances and detecting an object in the captured image as a cliff when cliff when a relation between the at least two compared determined distances complies with a predetermined increase criterion.

Description

This application is a U.S. National Phase application of PCT International Application No. PCT/EP2016/055547, filed Mar. 15, 2016, which is incorporated by reference herein.

TECHNICAL FIELD

The invention relates to a robotic cleaning device and a method at the robotic cleaning device of performing cliff detection along a surface over which the robotic cleaning device moves.

BACKGROUND

In many fields of technology, it is desirable to use robots with an autonomous behaviour such that they freely can move around a space without colliding with possible obstacles.

Robotic vacuum cleaners are known in the art, which are equipped with drive means in the form of a motor for moving the cleaner across a surface to be cleaned. The robotic vacuum cleaners are further equipped with intelligence in the form of microprocessor(s) and navigation means for causing an autonomous behaviour such that the robotic vacuum cleaners freely can move around and clean a surface in the form of e.g. a floor. Thus, these prior art robotic vacuum cleaners have the capability of more or less autonomously vacuum clean a room in which objects such as tables and chairs and other obstacles such as walls and stairs are located.

A particularly problematic obstacle encountered is a cliff, typically in the form of a stair leading down to a lower floor. If such a cliff is not detected by the robotic cleaner, there is a risk that the robot drops off a ledge, falls down the cliff and becomes permanently damaged.

WO 02/101477 discloses a robotic cleaner for performing autonomous cleaning of a surface. The robotic cleaner is equipped with reflective infrared (IR) proximity cliff sensors arranged at an underside of the robot and directed to observe the floor over which the robot moves. These cliff sensors emit IR beams and detect IR beams reflected from the surface over which the robot moves and upon encountering a cliff, no IR beam is reflected and the robot deducts that a cliff has been encountered.

However, a problem with this approach is that the robot detects the cliff as it is encountered, which renders it difficult for the robot to plan a cleaning path in advance to be followed for effectively cleaning the surface.

SUMMARY

An object of the present invention is to solve, or at least mitigate, this problem in the art and to provide an improved method at a robotic cleaning device of performing cliff detection.

This object is attained in a first aspect of the present invention by a method for a robotic cleaning device of performing cliff detection along a surface over which the robotic cleaning device moves. The method comprises illuminating the surface with at least one light source, capturing an image of the surface, detecting at least one illuminated section in the captured image, and determining distance to objects in the at least one illuminated section of the captured image. Further the method comprises comparing at least two of the determined distances and detecting an object in the captured image as a cliff when a relation between the at least two compared determined distances complies with a predetermined increase criterion.

This object is attained in a second aspect of the present invention by a robotic cleaning device comprising, a propulsion system arranged to move the robotic cleaning device over a surface to be cleaned, at least one light source arranged to illuminate the surface, a camera device arranged to capture an image of the surface, and a controller configured to control the propulsion system to move the robotic cleaning device and to control the camera device to capture the image. The controller is further configured to detect at least one illuminated section in the captured image, determine distance to objects in the at least one illuminated section of the captured image, compare at least two of the determined distances; and to detect an object in the captured image as a cliff when a relation between the at least two compared determined distances complies with a predetermined increase criterion.

Hence, the robotic cleaning device uses a light source illuminating the surface over which it moves while capturing images of the illuminated surface with its camera. By detecting illuminated objects captured in the images, it is possible for the robotic cleaning device to autonomously travel around the surface for performing a cleaning operation.

Now, upon encountering, e.g., a doorway where a stair leads down to a lower floor, a part of the light of the light source will fall onto the floor and ledge indicating the downwards-leading stair, while another part of the light falling beyond the ledge will impinge on for instance a wall of a stairwell leading down to the lower floor. This will cause a discontinuity in an illuminated section of the captured image (corresponding to the illuminated surface).

By having the robotic cleaning device determine distances to objects being illuminated in the captured image, it may be determined whether a distance from one detected object to another unexpectedly increases. Since the illuminated section (in which the objects are detected) of the captured image is discontinued at the ledge indicating the stair, the distance from the robotic cleaning device to an object residing beyond the ledge—e.g. the stairwell wall—will radically increase as compared to a preceding object located at or close to the ledge—e.g. the floor.

By comparing the distances and concluding that a relation between the at least two compared determined distances complies with a predetermined increase criterion stipulating that the increase in distance to the object beyond the ledge is sufficiently great with respect to the distance to the object located at the ledge, it can advantageously be concluded that a cliff has been detected.

Advantageously, this cliff is detected long before the robotic device encounters the ledge.

In an embodiment, when comparing two or more distances, for instance a first distance d3 with a second distance d4, a predetermined threshold value T may be utilized to determine whether the increase in distance is sufficiently great. Thus, the increase is considered sufficiently great when a relation between the at least two compared determined distances d4, d3 complies with the predetermined increase criterion.

In a further embodiment, the predetermined increase criterion is that a difference between the at least two compared determined distances exceeds a predetermined threshold value. For example, it may be determined whether Δ_d=d4−d3≥T, where the threshold value T is appropriately selected. If so, the increase in distance is to considered sufficiently great, and thus that a cliff advantageously has been detected.

In an alternative embodiment, it may be determined whether d4/d3≥T for concluding that the increase in distance is sufficiently great. In yet an alternative, it may be determined whether d3/d4≤T. Other alternatives of determining whether the increase Δ_d in distance is sufficiently great can be envisaged.

In another embodiment, a representation of surroundings of the robotic cleaning device is created from detected objects of the captured image, and an indication of the detected cliff is added to the created representation.

To this end, a controller derives positional data of the robotic cleaning device with respect to the surface to be cleaned from the detected objects of the recorded images and the associated determined distances, generates a 3D representation of the surroundings from the derived positional data and controls the robotic cleaning device to move across the surface to be cleaned in accordance with the generated 3D representation and navigation information supplied to the robotic cleaning device such that the surface to be cleaned can be autonomously navigated by taking into account the generated 3D representation.

The 3D representation generated from the images recorded by the camera thus facilitates detection of obstacles in the form of walls, floor lamps, table legs, around which the robotic cleaning device must navigate as well as rugs, carpets, doorsteps, etc., that the robotic cleaning device must traverse. The robotic cleaning device is hence configured to learn about its environment or surroundings by operating/cleaning.

The robotic cleaning device will in this embodiment advantageously add an indication of detected cliff(s) to the created representation of its surroundings.

In still another embodiment, the robotic cleaning device plans a cleaning path to be traversed by taking into account the detected cliff.

Thus, by adding to the created representation an indication of the cliff, it is advantageously possible for the robotic cleaning device to plan a cleaning path to be traversed well in advance and further to move very close to, and along, a ledge of the cliff since it is included in the representation of the surroundings of the robotic cleaning device.

Further in contrast to prior art cliff detectors, which detect a cliff just shortly before approaching the cliff and therefore needs to move at an amble pace to reduce the risk of falling down the cliff, the invention advantageously facilitates approaching a cliff at a relatively high speed, since the robotic cleaning device knows in advance exactly where the cliff is located and adds, to the created representation, and indication thereof accordingly.

In yet another embodiment, when detecting at least one illuminated section in the captured image, the robotic cleaning device further detects that a discontinuity occurs in the illuminated section of the captured image, where the illuminated section ceases to occur in the image beyond the detected discontinuity. Further in this embodiment, upon determining distance to objects in the illuminated section of the captured image, a value is assigned to a distance being determined before the discontinuity such that the assigned value reflects a sufficiently great increase.

Hence, it may be envisaged, depending on how the laser beam falls into the above exemplified stairwell, that the illuminated section beyond the ledge of the stairwell will not be present at all in the image. This indicates that a cliff is present, and this embodiment will advantageously detect the cliff.

Preferred embodiment of the present invention will be described in the following.

Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to “a/an/the element, apparatus, component, means, step, etc.” are to be interpreted openly as referring to at least one instance of the element, apparatus, component, means, step, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is now described, by way of example, with reference to the accompanying drawings, in which:

FIG. 1 illustrates detection of objects on a surface over which the robotic cleaning device moves in accordance with an embodiment of the present invention;

FIG. 2a illustrates an image captured by the camera of the robotic device when in position P1 of FIG. 1 according to an embodiment;

FIG. 2b illustrates an image captured by the camera of the robotic device when in position P2 of FIG. 1 according to an embodiment;

FIG. 3a illustrates the image of FIG. 2a with feature data indicated representing detected objects according to an embodiment;

FIG. 3b illustrates the image of FIG. 2b with feature data indicated representing detected objects according to an embodiment;

FIG. 4 illustrates a flowchart of the method of detecting a cliff according to an embodiment;

FIG. 5 illustrates the image of FIG. 2b with feature data indicated representing detected objects in a further scenario according to an embodiment;

FIG. 6 illustrates movement of the robotic cleaning device as a result of cliff detection according to an embodiment;

FIG. 7 shows a view of a robotic cleaning device according to an embodiment; and

FIG. 8 shows a further view of a robotic cleaning device according to an embodiment.

DETAILED DESCRIPTION

The invention will now be described more fully hereinafter with reference to the accompanying drawings, in which certain embodiments of the invention are shown. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided by way of example so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Like numbers refer to like elements throughout the description.

The invention relates to robotic cleaning devices, or in other words, to automatic, self-propelled machines for cleaning a surface, e.g. a robotic vacuum cleaner, a robotic sweeper or a robotic floor washer. The robotic cleaning device according to the invention can be mains-operated and have a cord, be battery-operated or use any other kind of suitable energy source, for example solar energy.

FIG. 1 illustrates detection of objects on a surface over which the robotic cleaning device moves in accordance with an embodiment of the present invention.

In this particular exemplifying embodiment, the robotic device too uses one vertical line laser 127 for illuminating the surface (i.e. floor 31) over which it moves. However, any appropriate light source may be envisaged for illuminating the surface 31. Further, a smaller or greater part of the surface may be illuminated.

As can be seen in FIG. 1, the line laser 127 projects a laser beam 30a onto the floor 31 and a first wall 32 of a room to be cleaned, while the robotic device 100 uses its camera 123 to capture images of the illuminated surface.

FIG. 2a illustrates an image 40a captured by the camera 123 of the robotic device 100 when in position P1.

As can be seen, the laser beam 30a will fall onto the floor 31 and the wall 32 and cause corresponding illuminated section 30a′ to be present in the image. The location in the image 40a where the floor 31 meets the wall 32 is indicated with a dashed line only for illustrational purposes; the dashed line is not present in the captured image 40a. From the captured image 40a, the robotic device 100 detects that the laser beam 30a impinges on an obstacle 32, such as a wall, a sofa, a door or the like. By capturing a number of images, the robotic device 100 is capable of identifying the particular obstacle 32 with high reliability. In case a different type of light source would be used, it may even be envisaged that the illuminated section covers the entire captured image. An advantage of using the laser beam as exemplified in FIG. 1 is that a less amount of image data need to be processed by the robotic device 100.

Now, again with reference to FIG. 1, when the robotic cleaning device 100 moves into position P2, it encounters a doorway 33 where a stair leads down to a lower floor. Again, the robotic device 100 captures an image of the illuminated surface.

FIG. 2b illustrates the image 40b captured by the camera 123 of the robotic device too when in position P2.

In this image, the laser beam 30b will fall onto the floor 31 and ledge 34 indicated with a dashed line in the image 40b for illustrational purposes. Further, the part of the laser beam 30b falling beyond the ledge 34 will incide on e.g. a wall of a stairwell leading down to the lower floor, which is indicated in the image 40b as a discontinuity in illuminated section 30b′ corresponding to the laser beam. It may alternatively be envisaged, depending on how the laser beam 30b falls into the stairwell, that the illuminated section 30b′ beyond the ledge 34 will not be present at all in the image 40b. Either way, the illuminated section 30b′ is discontinued at the ledge 34.

Hence, in contrast to prior art cliff detectors, the robotic cleaning device 100 is advantageously capable of detecting a cliff well in advance of actually encountering the ledge 34 indicating location of the cliff, since the discontinuity in the illuminated section 30b′ representing the projected laser beam indicates that a ledge 34 is encountered.

To this end, reference is made to FIGS. 3a and 3b showing the images 40a and 40b which previously was shown in FIGS. 2a and 2b respectively, but further with feature data indicated in the images representing detected objects. As can be seen, on each detected illuminated section 30a′, 30b in the respective image 40a and 40b, five feature points FP1-FP5 are illustrated representing objects detected in the images.

Hence, first with reference to FIG. 3a, the camera 123 of the robotic device 100 is controlled to capture the image 40a when in position P1 of FIG. 1. Again, the laser beam 30a will fall onto the floor 31 and the wall 32 and cause corresponding illuminated section 30a′ to be present in the image 40a.

As can be seen, the robotic device too determines distances to objects as represented by the feature data/points FP1-FP5 along the detected laser line 30a′ in the image 40a. Hence, from the first feature point FP1 to the second feature point FP2, the distance is d1; from the second feature point FP2 to the third point FP3, the distance is d2; and from the third feature point FP3 to the fourth point FP4, the distance is d3. As can be seen, in this particular example, the distance d1, d2, d3 between each pair of feature points is about the same; no radical change in distance is detected. The robotic device too will thus conclude that the feature points FP1-FP4 represent a floor 31.

Further, the distance from the robotic device 100 up to the fourth feature point FP4, i.e. the accumulated distances d1+d2+d3, is more or less the same as the distance from the robotic device 100 up to the fifth feature point FP5. Hence, the fifth feature FP5 will be considered by the robotic device too to represent an object in the form of a structure rising up from the floor 31, since FP4 and FP5 are located in the same plane. That is, the distance between the fourth feature point FP4 and the fifth feature point FP5 is practically zero. By detecting a further number of points along the illuminated section 30a (each being on the same distance from the robotic device 100), it may be concluded that the object indeed is a wall 32 due to its detected height, and not for instance a sofa or a chair.

Now, with reference to FIG. 3b, where the camera 123 of the robotic device too is controlled to capture the image 40b when in position P2 of FIG. 1. Again, the laser beam 30b will fall onto the floor 31, the ledge 34 and further into the stairwell of the stair leading down to the lower floor the wall 32 and cause corresponding illuminated to section 30b′ to be present in the image 40b.

Similar to the scenario described with reference to FIG. 3a, the robotic device 100 determines distances to objects as represented by the feature data/points FP1-Fp5 along the detected laser line 30b in the image 40b. Hence, from the first feature point FP1 to the second feature point FP2, the distance is d1; from the second feature point FP2 to the third point FP3, the distance is d2; and from the third feature point FP3 to the fourth point FP4, the distance is d3. As can be seen, in this particular example, the distance d1, d2, d3 between each pair of feature points is about the same; no radical change in distance is detected.

However, in this embodiment, the illuminated section 30b, is discontinued at the ledge 34. As a result, the distance d4 between the fourth feature point FP4 and the fifth feature point FP5 increases as compared to one or more of the previously determined distances. If the increase in distance d4 up to the fifth feature point FP5 is sufficiently great as compared to one or more of the previously determined distances, the feature point FP5 will be considered to represent a cliff. Advantageously, this cliff is detected long before the robotic device 100 encounters the ledge 34.

In an embodiment, when comparing two or more distances, for instance the third distance d3 with the fourth distance d4, a predetermined threshold value T may be utilized to determine whether the increase in distance is sufficiently great. Thus, the increase is considered sufficiently great when a relation between the at least two compared determined distances d4, d3 complies with the predetermined increase criterion.

For example, it may be determined whether Δ_d=d4−d3≥T, where the threshold value T is appropriately selected. If so, the increase in distance is considered sufficiently great, and thus that a cliff advantageously has been detected.

Alternatively, it may be determined whether d4/d3≥T for concluding that the increase in distance is sufficiently great. In yet an alternative, it may be determined whether d3/d4≤T. Other alternatives of determining whether the increase Δ_d in distance is to sufficiently great can be envisaged.

FIG. 4 illustrates a flowchart of the method of detecting a cliff according to an embodiment. In step S101, the robotic device 19 uses a light source, for instance a vertical line laser 127 as previously discussed (or any other appropriate light source), for illuminating a surface 31 over which the robotic device 100 moves.

In step S102, the robotic device 100 uses its camera 123 to capture images 40a, 40b of the illuminated surface.

The laser beam 30a, 30b will fall onto the floor 31/wall 32 and the floor 31/doorway 33, respectively, and cause corresponding illuminated section 30a′, 30b′ to be present in the images 40a, 40b, which illuminated section is detected by the robotic device too in step S103.

In step S104, the robotic device too determines, as previously has been discussed in detail with reference to FIG. 3b, distances to objects as represented by the feature points FP1-Fp5 along the detected laser line 30b in the image 40b.

Thereafter, in step S105, the determined distances to illuminated objects in the image 40 are compared to determine whether a sufficiently great increase Δ_d has occurred in the compared distances. For instance, if it is determined that Δ_d=d4−d3≥T, where the threshold value T is appropriately selected, possibly depending on the particular application and surroundings.

In step S106, if the increase in distance d4 up to the fifth feature point FP5 is sufficiently great as compared to the distance d3 up to the fourth feature point FP4, the feature point FP5 will be considered to represent a cliff. Advantageously, this cliff is detected long before the robotic device 100 encounters the ledge 34.

With reference to FIG. 5, in a further embodiment, if the laser beam 30b falls into the stairway in an angle such that it is not captured in the image 40b, i.e. if the part of the laser beam falling beyond the ledge 34 falls outside of the captured image 40b, the distance to any point along such laser beam 30b will be considered to be infinite (or in practice be given a great numerical value).

Hence, if the detected illuminated section 30b′ in the image 40b is discontinued, and the illuminated section 30b′ ceases to occur after the discontinuity, a value is assigned to a distance being determined before the discontinuity occurring—for instance to distance d3—such that the assigned value reflects an increase Δ_d sufficiently great for a cliff to advantageously be detected.

In a further embodiment, the camera 123 is controlled by a controller such as a microprocessor to capture and record images from which the controller creates a representation or layout of the surroundings that the robotic cleaning device 100 is operating in, by extracting feature points from the images representing detected objects and by measuring the distance from the robotic cleaning device 100 to these objects, while the robotic cleaning device 100 is moving across the surface to be cleaned. Thus, the controller derives positional data of the robotic cleaning device 100 with respect to the surface to be cleaned from the detected objects of the recorded images, generates a 3D representation of the surroundings from the derived positional data and controls driving motors to move the robotic cleaning device 100 across the surface to be cleaned in accordance with the generated 3D representation and navigation information supplied to the robotic cleaning device 100 such that the surface to be cleaned can be autonomously navigated by taking into account the generated 3D representation. Since the derived positional data will serve as a foundation for the navigation of the robotic cleaning device, it is important that the positioning is correct; the robotic device will otherwise navigate according to a “map” of its surroundings that is misleading.

The 3D representation generated from the images recorded by the camera 123 thus facilitates detection of obstacles in the form of walls, floor lamps, table legs, around which the robotic cleaning device must navigate as well as rugs, carpets, doorsteps, etc., that the robotic cleaning device 100 must traverse. The robotic cleaning device 100 is hence configured to learn about its environment or surroundings by operating/cleaning.

The robotic cleaning device 100 will advantageously add an indication of detected cliff(s) to the created representation of its surroundings. Thus, by adding to the created representation an indication of the ledge 34, it is possible for the robotic cleaning device to plan a cleaning path to be traversed well in advance and further to move very close to, and along, the ledge 34 since it is included in the representation of the surroundings of the robotic cleaning device 100.

This is illustrated in FIG. 6, where the robotic cleaning device moves from position P3 and advantageously knows well in advance where the doorway 33 and the ledge 34 is located. The robotic device too will thus move into position P4 and travel flush to ledge 34 (and even with a part of a body of the robotic device protruding out from the ledge 34).

Further in contrast to prior art cliff detectors, which detect a cliff just shortly before approaching the cliff and therefore needs to move at an amble pace to reduce the risk of falling down the cliff, it is possible with the invention to approach a cliff at a relatively high speed, since the robotic cleaning device too knows in advance exactly where the cliff is located and adds, to the created representation, and indication thereof accordingly.

Even though it is envisaged that the invention may be performed by any appropriate robotic cleaning device being equipped with sufficient processing intelligence, FIG. 7 shows a robotic cleaning device 100 according to an embodiment of the present invention in a bottom view, i.e. the bottom side of the robotic cleaning device is shown. The arrow indicates the forward direction of the robotic cleaning device 100 being illustrated in the form of a robotic vacuum cleaner.

The robotic cleaning device 100 comprises a main body 111 housing components such as a propulsion system comprising driving means in the form of two electric wheel motors 115a, 115b for enabling movement of the driving wheels 112, 113 such that the cleaning device can be moved over a surface to be cleaned. Each wheel motor 115a, 115b is capable of controlling the respective driving wheel 112, 113 to rotate independently of each other in order to move the robotic cleaning device 100 across the surface to be cleaned. A number of different driving wheel arrangements, as well as various wheel motor arrangements, can be envisaged. It should be noted that the robotic cleaning device may have any appropriate shape, such as a device having a more traditional circular-shaped main body, or a triangular-shaped main body. As an alternative, a track propulsion system may be used or even a hovercraft propulsion system. The propulsion system may further be arranged to cause the robotic cleaning device too to perform any one or more of a yaw, pitch, translation or roll movement.

A controller 116 such as a microprocessor controls the wheel motors 115a, 115b to rotate the driving wheels 112, 113 as required in view of information received from an obstacle detecting device (not shown in FIG. 7) for detecting obstacles in the form of walls, floor lamps, table legs, around which the robotic cleaning device must navigate. The obstacle detecting device may be embodied in the form of a 3D sensor system registering its surroundings, implemented by means of e.g. a 3D camera, a camera in combination with lasers, a laser scanner, etc. for detecting obstacles and communicating information about any detected obstacle to the microprocessor 116. The microprocessor 116 communicates with the wheel motors 115a, 115b to control movement of the wheels 112, 113 in accordance with information provided by the obstacle detecting device such that the robotic cleaning device 100 can move as desired across the surface to be cleaned.

Further, the main body 111 may optionally be arranged with a cleaning member 117 for removing debris and dust from the surface to be cleaned in the form of a rotatable brush roll arranged in an opening 118 at the bottom of the robotic cleaner 100. Thus, the rotatable brush roll 117 is arranged along a horizontal axis in the opening 118 to enhance the dust and debris collecting properties of the cleaning device 100. In order to rotate the brush roll 117, a brush roll motor 119 is operatively coupled to the brush roll to control its rotation in line with instructions received from the controller 116.

Moreover, the main body 111 of the robotic cleaner 100 comprises a suction fan 20 creating an air flow for transporting debris to a dust bag or cyclone arrangement (not shown) housed in the main body via the opening 118 in the bottom side of the main body 111. The suction fan 120 is driven by a fan motor 121 communicatively connected to the controller 116 from which the fan motor 121 receives instructions for controlling the suction fan 120. It should be noted that a robotic cleaning device having either one of the rotatable brush roll 117 and the suction fan 20 for transporting debris to the dust bag can be envisaged. A combination of the two will however enhance the debris-removing capabilities of the robotic cleaning device 100.

The main body 111 or the robotic cleaning device too is further equipped with an inertia measurement unit (IMU) 124, such as e.g. a gyroscope and/or an accelerometer and/or a magnetometer or any other appropriate device for measuring displacement of the robotic cleaning device too with respect to a reference position, in the form of e.g. orientation, rotational velocity, gravitational forces, etc. A three-axis gyroscope is capable of measuring rotational velocity in a roll, pitch and yaw movement of the robotic cleaning device 100. A three-axis accelerometer is capable of measuring acceleration in all directions, which is mainly used to determine whether the robotic cleaning device is bumped or lifted or if it is stuck (i.e. not moving even though the wheels are turning). The robotic cleaning device 100 further comprises encoders (not shown in FIG. 7) on each drive wheel 112, 113 which generate pulses when the wheels turn. The encoders may for instance be magnetic or optical. By counting the pulses at the controller 116, the speed of each wheel 112, 113 can be determined. By combining wheel speed readings with gyroscope information, the controller 116 can perform so called dead reckoning to determine position and heading of the cleaning device 100.

The main body 111 may further be arranged with a rotating side brush 114 adjacent to the opening 118, the rotation of which could be controlled by the drive motors 115a, 115b, the brush roll motor 119, or alternatively a separate side brush motor (not shown). Advantageously, the rotating side brush 114 sweeps debris and dust such from the surface to be cleaned such that the debris ends up under the main body 111 at the opening 118 and thus can be transported to a dust chamber of the robotic cleaning device. Further advantageous is that the reach of the robotic cleaning device 100 will be improved, and e.g. corners and areas where a floor meets a wall are much more effectively cleaned. As is illustrated in FIG. 7, the rotating side brush 114 rotates in a direction such that it sweeps debris towards the opening 118 such that the suction fan 20 can transport the debris to a dust chamber. The robotic cleaning device too may comprise two rotating side brushes arranged laterally on each side of, and adjacent to, the opening 118.

With further reference to FIG. 7, the controller/processing unit 116 embodied in the form of one or more microprocessors is arranged to execute a computer program 125 downloaded to a suitable storage medium 126 associated with the microprocessor, such as a Random Access Memory (RAM), a Flash memory or a hard disk drive. The controller 116 is arranged to carry out a method according to embodiments of the present invention when the appropriate computer program 125 comprising computer-executable instructions is downloaded to the storage medium 126 and executed by the controller 116. The storage medium 126 may also be a computer program product comprising the computer program 125. Alternatively, the computer program 125 may be transferred to the storage medium 126 by means of a suitable computer program product, such as a digital versatile disc (DVD), compact disc (CD) or a memory stick. As a further alternative, the computer program 125 may be downloaded to the storage medium 126 over a wired or wireless network. The controller 116 may alternatively be embodied in the form of a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a complex programmable logic device (CPLD), etc.

FIG. 8 shows a front view of the robotic cleaning device too of FIG. 7 in an embodiment of the present invention illustrating the previously mentioned obstacle detecting device in the form of a 3D sensor system comprising at least a camera 123 and a first and a second line laser 127, 128, which may be horizontally or vertically oriented line lasers. Further shown is the controller 116, the main body 111, the driving wheels 112, 113, and the rotatable brush roll 117 previously discussed with reference to FIG. 6. The controller 116 is operatively coupled to the camera 123 for recording images of a vicinity of the robotic cleaning device 100. The first and second line lasers 127, 128 may preferably be vertical line lasers and are arranged lateral of the camera 123 and configured to illuminate a height and a width that is greater than the height and width of the robotic cleaning device 100. Further, the angle of the field of view of the camera 123 is preferably smaller than the space illuminated by the first and second line lasers 127, 128. The camera 123 is controlled by the controller 116 to capture and record a plurality of images per second. Data from the images is extracted by the controller 116 and the data is typically saved in the memory 126 along with the computer program 125.

The first and second line lasers 127, 128 are typically arranged on a respective side of the camera 123 along an axis being perpendicular to an optical axis of the camera. Further, the line lasers 127, 128 are directed such that their respective laser beams intersect within the field of view of the camera 123. Typically, the intersection coincides with the optical axis of the camera 123.

The first and second line laser 127, 128 are configured to scan, preferably in a vertical orientation, the vicinity of the robotic cleaning device 100, normally in the direction of movement of the robotic cleaning device 100. The first and second line lasers 127, 128 are configured to send out laser beams, which illuminate furniture, walls and other objects of e.g. a room to be cleaned. The camera 123 is controlled by the controller 116 to capture and record images from which the controller 116 creates a representation or layout of the surroundings that the robotic cleaning device 100 is operating in, by extracting features from the images and by measuring the distance covered by the robotic cleaning device 100, while the robotic cleaning device too is moving across the surface to be cleaned. Thus, the controller 16 derives positional data of the robotic cleaning device too with respect to the surface to be cleaned from the recorded images, generates a 3D representation of the surroundings from the derived positional data and controls the driving motors 115a, 115b to move the robotic cleaning device across the surface to be cleaned in accordance with the generated 3D representation and navigation information supplied to the robotic cleaning device 100 such that the surface to be cleaned can be navigated by taking into account the generated 3D representation. Since the derived positional data will serve as a foundation for the navigation of the robotic cleaning device, it is important that the positioning is correct; the robotic device will otherwise navigate according to a “map” of its surroundings that is misleading.

The 3D representation generated from the images recorded by the 3D sensor system thus facilitates detection of obstacles in the form of walls, floor lamps, table legs, around which the robotic cleaning device must navigate as well as rugs, carpets, doorsteps, etc., that the robotic cleaning device too must traverse. The robotic cleaning device 100 is hence configured to learn about its environment or surroundings by operating/cleaning.

Hence, the 3D sensor system comprising the camera 123 and the first and second vertical line lasers 127, 128 is arranged to record images of a vicinity of the robotic cleaning from which objects/obstacles may be detected. The controller 116 is capable of positioning the robotic cleaning device 100 with respect to the detected obstacles and hence a surface to be cleaned by deriving positional data from the recorded images. From the positioning, the controller 116 controls movement of the robotic cleaning device 100 by means of controlling the wheels 112, 113 via the wheel drive motors 115a, 115b, across the surface to be cleaned.

The derived positional data facilitates control of the movement of the robotic cleaning device 100 such that cleaning device can be navigated to move very close to an object, and to move closely around the object to remove debris from the surface on which the object is located. Hence, the derived positional data is utilized to move flush against the object, being e.g. a chair, a table, a sofa, a thick rug or a wall. Typically, the controller 116 continuously generates and transfers control signals to the drive wheels 112, 113 via the drive motors 115a, 115b such that the robotic cleaning device 100 is navigated close to the object.

It should further be noted that while the embodiments of the invention has been discussed in the context of using a camera and one or two line lasers for illuminating a surface over which the robotic cleaning device 100 moves, it would further be possible to use known 3D sensors utilizing time of flight measurements of an image being completely illuminated. With such a time of flight 3D sensor, the distance in a captured image would be determined for each pixel and distances to detected objects may be determined in line with the above.

The invention has mainly been described above with reference to a few embodiments. However, as is readily appreciated by a person skilled in the art, other embodiments than the ones disclosed above are equally possible within the scope of the invention, as defined by the appended patent claims.

Claims

1. A method for a robotic cleaning device of performing cliff detection along a surface over which the robotic cleaning device moves, the method comprising: illuminating the surface with a corresponding line laser;capturing an image of the surface;detecting at least one illuminated section illuminated by the corresponding line laser in the captured image;determining distances between successive measurement points along the corresponding line laser, the successive measurement points corresponding to multiple points in front of the robotic cleaning device that are illuminated by the corresponding line laser in the captured image;comparing at least two of the determined distances to one another; anddetecting an object in the captured image as a cliff when a difference between the at least two compared determined distances exceeds a threshold value.

2. A non-transitory computer readable medium, the computer readable medium having a computer program stored thereon, that when executed by a processor, instructs the processor to perform the steps of: illuminating the surface with a corresponding line laser;capturing an image of the surface;detecting at least one illuminated section illuminated by the corresponding line laser in the captured image;determining distances between successive measurement points along the corresponding line laser, the successive measurement points corresponding to multiple points in front of the robotic cleaning device that are illuminated by the corresponding line laser in the captured image;comparing at least two of the determined distances to one another; anddetecting an object in the captured image as a cliff when a difference between the at least two compared determined distances exceeds a threshold value.

3. The method of claim 1, further comprising: creating a representation of surroundings of the robotic cleaning device from detected objects of the captured image; andadding an indication of the detected cliff to the created representation.

4. The method of claim 1, further comprising planning a cleaning path to be traversed by taking into account the detected cliff.

5. The method of claim 1, wherein the step of detecting at least one illuminated section in the captured image further comprises detecting that a discontinuity occurs in the at least one illuminated section in the captured image, where the at least one illuminated section ceases to occur in the image beyond the detected discontinuity; andthe step of determining distance to objects in the at least one illuminated section of the captured image further comprises assigning a value to a distance being determined before the discontinuity such that the assigned value reflects a sufficiently great increase.

6. A robotic cleaning device comprising: a propulsion system arranged to move the robotic cleaning device over a surface to be cleaned;a corresponding line laser arranged to illuminate the surface;a camera arranged to capture an image of the surface; anda controller configured to control the propulsion system to move the robotic cleaning device and to control the camera to capture the image; wherein the controller further is configured to: detect at least one illuminated section illuminated by the corresponding line laser in the captured image;determine distances between successive measurement points along the corresponding line laser, the successive measurement points corresponding to multiple points in front of the robotic cleaning device that are illuminated by the corresponding line laser in the captured image;compare at least two of the determined distances to one another; anddetect an object in the captured image as a cliff when a difference between the at least two compared determined distances exceeds a threshold value.

7. The robotic cleaning device of claim 6, wherein the line laser is a vertical line laser.

8. The robotic cleaning device of claim 6, the controller further being configured to: create a representation of surroundings of the robotic cleaning device from detected objects of the captured image; andadd an indication of the detected cliff to the created representation.

9. The robotic cleaning device of claim 6, the controller further being configured to plan a cleaning path to be traversed by taking into account the detected cliff.

10. The robotic cleaning device of claim 6, the controller further being configured to: when detecting at least one illuminated section in the captured image, detect that a discontinuity occurs in the at least one illuminated section in the capture image, where the at least one illuminated section cease to occur in the image beyond the detected discontinuity; andwhen determining distance to objects in the at least one illuminated section of the captured image, assign a value to a distance being determined before the discontinuity such that the assigned value reflects a sufficiently great increase.

11. The robotic cleaning device of claim 7, further comprising another vertical line laser arranged to illuminate the surface.

12. The robotic cleaning device of claim 11, wherein the vertical line laser and the another vertical line laser are arranged on respective sides of the camera along an axis being perpendicular to an optical axis of the camera.