SYSTEM AND DEVICE SUITABLE FOR Facilitating WAGON TRANSPORT, AND ASSOCIATED PROCESSING METHOD

Abstract

A processing procedure (300) is provided that is suitable for facilitating cart transport. The processing procedure (300) can include an initialization step (302) in which a status estimate is determined in connection with a cart status associated with a cart, wherein the cart has a body and several wheels coupled to the body, and an inference step (306) in which an update of the overall status associated with the cart can be derived, wherein the update of the overall status is associated with a cost function. The cost function can include a parameter for longitudinal position errors with respect to the cart body, a parameter for lateral position errors between the corners of the cart body, if detectable, and a positioning parameter corresponding to the positioning of the wheels with respect to the status estimate.

Description

Field of invention

[0001] The present disclosure relates generally to a system and / or device suitable for facilitating wagon transport. The present disclosure further relates to a processing method that can be combined with the system and / or device. background

[0002] Automated guided vehicles (AGVs) can be useful for wagon detection and subsequent wagon retrieval. For example, an AGV can be primarily used to pick up one or more wagons and move them to one or more other desired locations (that is, wagon transport).

[0003] AGVs are typically equipped with one or more sensors to facilitate detection. Furthermore, most carts can be connected with a fixed shape / size, so an approach based on matching predefined templates may be used (see, for example, Bostelman, Roger, Roger Bostelman and Tsai Hong. Review of Research for Docking Automatic Guided Vehicles and Mobile Robots. US-Department of Commerce, National Institute of Standards and Technology, 2016).

[0004] In this respect, it is understood that the approach of matching predefined templates is an example of a conventional technique for identifying one or more wagons in order to facilitate wagon transport.

[0005] This disclosure acknowledges that conventional techniques for facilitating wagon transport still need further improvement. Brief description of the invention

[0006] According to one aspect of the disclosure, a processing method is provided that may be suitable, for example, for facilitating the transport of vehicles. In one embodiment, the processing method may, for example, include an initialization step and an inference step. In another embodiment, the processing method may, for example, further include a preprocessing step.

[0007] Regarding the initialization step, a status estimate can be determined in conjunction with a cart status. The cart status can be linked to a cart. The cart can, for example, consist of a cart body and multiple wheels. The multiple wheels can be coupled to the cart body. Generally, the wheels can be designed, for example, to facilitate the movement of the cart body.

[0008] Regarding the inference step, an update of the overall status can be derived, which can be linked to the cart. This update can be associated with a cost function. The cost function might include, for example, a parameter for longitudinal position errors with respect to the cart body, a parameter for lateral position errors between the corners of the cart body (if detectable), and a positioning parameter corresponding to the positioning of the wheels relative to the status estimate.

[0009] Regarding the preprocessing step, the positions of the wheels and / or the orientations of the wheels can be determined, for example, based on the recorded data.

[0010] Accordingly, the present disclosure considers a sensor fusion approach that may be based, for example, on a combination of detecting one or more cart wheels and detecting the cart body, according to one embodiment of the disclosure. This can be helpful, for example, in cart docking (referred to, for example, as "car cart pickup"), in conjunction with carts equipped with steering wheels. According to one embodiment of the disclosure, it is taken into account that, in an example of a cart with steering wheels (which may be associated with arbitrary rotation about a vertical axis), ambiguities may arise when the cart position is reset from the detection of one or more (steering) wheels.

[0011] The above approach takes into account that at least one robust method (which, for example, addresses the aforementioned ambiguities) and / or one efficient method (which, for example, mitigates the need for pickup / detection aids) can be provided / facilitated for vehicle detection and / or pickup according to an embodiment of the disclosure.

[0012] The present disclosure further includes a computer program which may contain instructions which, when the program is executed by a computer, cause the computer to perform the initialization step, the preprocessing step and / or the inference step, as discussed with reference to the processing procedure.

[0013] However, the present disclosure further includes a computer-readable storage medium (not shown) containing data stored therein that constitutes computer-executable software, wherein the software includes instructions which, when executed by the computer, perform the initialization step, the preprocessing step and / or the inference step, as discussed with reference to the processing procedure.

[0014] The advantageous aspects of the processing method described above in the present disclosure can also be applied analogously to all aspects of a device described below in the present disclosure. Likewise, all advantageous aspects of the device described below in the disclosure can also be applied analogously to all aspects of the processing method described above in the disclosure.

[0015] According to one aspect of the disclosure, a device is provided which may, for example, be suitable for facilitating cart transport. The device may, for example, be combined with the processing method according to one embodiment of the disclosure.

[0016] The device can, for example, include a first module, a second module and a third module, according to one embodiment of the disclosure.

[0017] In one embodiment, the first module (which corresponds, for example, to a receiver) can be configured to receive at least one input signal. The second module (which corresponds, for example, to a processor) can be configured to process the one or more input signals according to the processing method discussed above in order to generate at least one output signal. The third module (which corresponds, for example, to a transmitter) can be configured to communicate the one or more output signals. The one or more output signals can be communicated, for example, to a robot (for example, an AGV) and / or one or more devices. The one or more output signals can be used, for example, to facilitate cart transport (for example, for cart detection and / or cart retrieval).

[0018] Accordingly, the present disclosure considers a sensor fusion approach that may be based, for example, on a combination of detecting one or more cart wheels and detecting the cart body, according to one embodiment of the disclosure. This can be helpful, for example, in cart docking (referred to, for example, as "car cart pickup"), in conjunction with carts equipped with steering wheels. According to one embodiment of the disclosure, it is taken into account that, in an example of a cart with steering wheels (which may be associated with arbitrary rotation about a vertical axis), ambiguities may arise when the cart position is reset from the detection of one or more (steering) wheels.

[0019] The above approach takes into account that at least one robust method (which, for example, addresses the aforementioned ambiguities) and / or one efficient method (which, for example, mitigates the need for pickup / detection aids) can be provided / facilitated for vehicle detection and / or pickup according to an embodiment of the disclosure. Brief description of the drawings

[0020] Embodiments of the disclosure are described below with reference to the following drawings, in which the following applies: Fig. Figure 1 shows a system which may include at least one device according to an embodiment of the disclosure; Fig. 2 shows the device of Fig. 1 according to one embodiment of the disclosure in more detail; and Fig. Figure 3 shows a processing method in conjunction with the system of Fig. 1 according to one embodiment of the disclosure. Detailed description

[0021] This disclosure takes into account that wagon transport can be combined with wagon identification and / or wagon collection (i.e., wagon identification and / or wagon collection).

[0022] An automated guided vehicle (AGV) can, for example, carry multiple sensors to facilitate detection (e.g., vehicle detection). The sensors can include, for example, one or more light detection and distance measurement (Lidar) sensors and / or one or more depth sensors. An example of a Lidar sensor could be a 2D (two-dimensional) Lidar camera, and an example of a depth sensor could be a depth camera. In a specific example, according to one embodiment of the disclosure, an AGV can carry at least one 2D Lidar camera and at least one depth camera.

[0023] Based on wagon detection, the AGV can be trained to perform one or more tasks related to picking up one or more wagons. After wagon pickup, the AGV can be trained to move (for example, transport) the one or more wagons from one location to another (desired) location. Wagon pickup (for example, referred to as "wagon docking") can, for example, be based on an initial approach phase (for example, when the AGV first approaches a general location where a wagon is parked) and a final alignment phase (for example, after the initial approach phase, in which the AGV performs a fine-position alignment task with respect to the wagon in order to dock the wagon to the AGV) according to one embodiment of the disclosure.

[0024] This disclosure takes into account that in certain situations, information regarding a precise 3D model of a cart may not be available in advance. It further acknowledges that it may be beneficial to facilitate cart retrieval without requiring retrieval / detection aids (for example, retrofitting one or more alignment markers, retrofitting special lockable wheels, and / or providing cart docking stations). It also acknowledges that retrieval / detection aids (such as retrofitting) may increase operating costs. Finally, it considers that for a cart with steering wheels (which may be associated with arbitrary rotation about a vertical axis), ambiguities may arise when the cart's position is reset based on the detection of one or more (steering) wheels.

[0025] Accordingly, the present disclosure takes into account a sensor fusion approach that may, for example, be based on a combination of detecting one or more wagon wheels and detecting the wagon body, according to one embodiment of the disclosure. This can be helpful, for example, in wagon docking (referred to, for example, as "wagon pickup"), in conjunction with wagons equipped with swivel casters.

[0026] The above approach takes into account that at least one robust method (which, for example, addresses the aforementioned ambiguities) and / or one efficient method (which, for example, mitigates the need for pickup / detection aids) can be provided / facilitated for vehicle detection and / or pickup according to an embodiment of the disclosure.

[0027] The foregoing is based on Fig. 1 to Fig. 3 discussed in more detail below.

[0028] Referring to Fig. Figure 1 shows System 100 according to one embodiment of the disclosure. System 100 can, for example, be suitable for facilitating wagon transport (can, for example, be combined with wagon identification and / or wagon collection), according to one embodiment of the disclosure.

[0029] As shown, the system 100 can include one or more devices 102, at least one device 104 and optionally a communication network 106 according to an embodiment of the disclosure.

[0030] The one or more devices 102 can be coupled with the one or more devices 104. In particular, according to one embodiment of the disclosure, the one or more devices 102 can be coupled with the one or more devices 104 via the communication network 106.

[0031] In one embodiment, the one or more devices 102 can be coupled to the communication network 106, and the one or more devices 104 can be coupled to the communication network 106. The coupling can be effected by wired coupling and / or wireless coupling. According to one embodiment of the disclosure, the one or more devices 102 can generally be configured to communicate with the one or more devices 104 via the communication network 106.

[0032] In general, according to one embodiment of the disclosure, the one or more devices 102 can be configured to receive one or more input signals and to process the one or more input signals in order to generate / derive one or more output signals. Furthermore, according to one embodiment of the disclosure, the one or more devices 104 can, for example, be configured to generate the one or more input signals and / or to communicate the one or more input signals to the one or more devices 102.

[0033] The one or more devices 102 can, for example, be configured to process the one or more input signals in order to generate / derive the one or more output signals, according to one embodiment of the disclosure. In one embodiment, the one or more devices 102 can, for example, be carried by an AGV. In another embodiment, the one or more devices 102 can, for example, be arranged remotely with respect to an AGV (for example, be in remote communication with an AGV and not be carried by the AGV). In yet another embodiment, a section of the one or more devices 102 can, for example, be carried by an AGV, and another section of the one or more devices 102 can, for example, be arranged remotely with respect to the AGV.The one or more devices 102 are described in accordance with one embodiment of the disclosure with reference to . Fig. 2 discussed in more detail.

[0034] The one or more devices 104 can, for example, be configured to generate the one or more input signals and / or to communicate the one or more input signals, according to one embodiment of the disclosure. For example, the one or more input signals can be communicated from the one or more devices 104 to the one or more devices 102. In one example, a device 104 can be connected to, correspond to, or include one or more sensors (for example, a 2D lidar camera and / or a depth camera). In one embodiment, the one or more devices 104 can, for example, be carried by an AGV. In another embodiment, the one or more devices 104 can, for example, be located remotely with respect to an AGV (for example, be in remote communication with an AGV and not be carried by the AGV).In yet another embodiment, a section of the one or more devices 104 can, for example, be carried by an AGV, and another section of the one or more devices 104 can, for example, be arranged remotely in relation to the AGV.

[0035] The communication network 106 can, for example, correspond to an internet communication network, a wired communication network, a wireless communication network, or any combination thereof. Communication (that is, between the one or more devices 102 and the one or more devices 104) via the communication network 106 can take place through wired and / or wireless communication.

[0036] In a general example, one or more input signals (which may include, for example, 2D lidar data and / or depth camera data) can be communicated by one or more devices 104 and received by one or more devices 102 for processing in order to generate one or more output signals that can be communicated by one or more devices 104. The one or more output signals can, for example, correspond to one or more control signals that can be used, for example, for the navigation of an AGV to facilitate cart transport (which can be linked, for example, to cart detection and / or cart pickup), according to one embodiment of the disclosure.

[0037] The one or more of the aforementioned devices 102 are described below with reference to Fig. 2 discussed in more detail.

[0038] Referring to Fig. 2 A device 102 is shown in more detail in connection with exemplary implementation 200 according to an embodiment of the disclosure.

[0039] In exemplary implementation 200, the device 102 can correspond to electronic module 200a, which, for example, may be suitable for performing one or more processing tasks according to an embodiment of the disclosure.

[0040] The electronic module 200a can, for example, contain housing 200b. Furthermore, the electronic module 200a can, for example, carry first module 202, second module 204, third module 206, or any combination thereof.

[0041] In one embodiment, the electronic module 200a can carry a first module 202, a second module 204, and / or a third module 206. In a specific example, according to one embodiment of the disclosure, the electronic module 200a can carry a first module 202, a second module 204, and a third module 206.

[0042] In this sense, it is understood that in one embodiment the housing 200b can be shaped and dimensioned such that it can support the first module 202, the second module 204, the third module 206 or any combination thereof.

[0043] The first module 202 can be coupled to the second module 204 and / or the third module 206. The second module 204 can be coupled to the first module 202 and / or the third module 206. The third module 206 can be coupled to the first module 202 and / or the second module 204. In one embodiment of the disclosure, the first module 202 can be coupled to the second module 204, and the second module 204 can be coupled to the third module 206. The coupling between the first module 202, the second module 204, and / or the third module 206 can be, for example, by wired and / or wireless coupling. According to one embodiment of the disclosure, the first module 202, the second module 204, and / or the third module 206 can correspond to a hardware-based module and / or a software-based module.

[0044] In one example, the first module 202 can correspond to a hardware-based receiver that can be configured to receive one or more input signals.

[0045] The second module 204 can, for example, correspond to a network-based / software-based and / or hardware-based (for example, a microprocessor) processing module, which can be trained to perform one or more processing tasks in conjunction with one or any combination of the following: • Initialization • Preprocessing • Inference-based processing

[0046] In particular, the second module 204 can, for example, be configured to process one or more received input signals by initialization, preprocessing and / or inference-based processing in such a way that one or more output signals are generated / derived according to an embodiment of the disclosure.

[0047] The third module 206 can, in one example, correspond to a hardware-based transmitter configured to communicate one or more output signals from the electronic module 200a, according to one embodiment of the disclosure. The one or more output signals can, for example, be communicated from the electronic module 200a to one or more devices 104 and / or one or more other devices 102, according to one embodiment of the disclosure.

[0048] The present disclosure takes into account the possibility that the first and second modules 202 / 204 may be integrated software-hardware-based modules (for example, an electronic part that may carry a software program / algorithm in conjunction with receiving and processing functions / an electronic module programmed to perform the receiving and processing functions). The present disclosure further takes into account the possibility that the first and third modules 202 / 206 may be integrated software-hardware-based modules (for example, an electronic part that may carry a software program / algorithm in conjunction with receiving and transmitting functions / an electronic module programmed to perform the receiving and transmitting functions).The present disclosure further takes into account the possibility that the first and third modules 202 / 206 may be integrated hardware modules (for example, hardware-based transceivers) capable of performing the functions of receiving and transmitting.

[0049] The exemplary implementation 200 above will now be discussed in more detail below with reference to an exemplary scenario according to an embodiment of the disclosure.

[0050] In the exemplary scenario, the second module 204 can, for example, correspond to a processor that is capable of processing the one or more received input signals at any time by initialization, preprocessing and / or inference-based processing (that is, performing one or more processing tasks) in order to generate one or more output signals according to an embodiment of the disclosure.

[0051] The aforementioned initialization, preprocessing, and inference-based processing are discussed below in succession, according to one embodiment of the disclosure.

[0052] With regard to initialization, the second module 204 can be configured to initialize a standard status estimate in connection with at least one cart (for example, a cart status associated with a cart). The standard status estimate (simply referred to as the "status estimate") of the cart status can, for example, include one or more parameters such as cart body position, cart size / shape, wheel size and caster offset, detectable element topology, position / rotation angle, or any combination thereof. In one example, the status estimate of the cart status can include parameters such as cart body position, cart size / shape, wheel size and caster offset, detectable element topology, and position / rotation angle, according to one embodiment of the disclosure.In another example, the status estimation of the vehicle status according to an embodiment of the disclosure may further include one or more other parameters that can specify the positions of the wheels.

[0053] In one embodiment, the carriage body position can be assumed to define the center point of a coordinate system. The carriage size / shape can be based on dimensions (for example, carriage body width, carriage body length, longitudinal wheel offset, and / or lateral wheel offset) associated with a carriage (for example, status_width, status_length, status_longitudinal_wheel_offset, status_lateral_wheel_offset). Furthermore, wheel size and caster offset can be based on dimensions (for example, width of one or more wheels, radius of one or more wheels, and / or caster offset) associated with one or more wheels of the carriage (for example, status_wheel_width, status_wheel_radius, status_caster_offset).Additionally, the detectable element topology can be based on a condition that all casters can be observed, or a condition that the areas associated with the casters can be observed (that is, each caster can be assigned to an area, such as a rectangular area in which the caster can lie), or a condition that at least 2 areas associated with 4 casters can be observed (for example, two rectangular areas in which a pair of wheels can lie in each rectangular area), or any combination thereof.Furthermore, the position / rotation angle can be based on a position / rotation angle associated with an AGV (for example, referred to as a "robot"), and such a position (for example, position in the "X" coordinate and position in the "Y" coordinate) / rotation angle can be expressed as a trajectory initialized by the odometry data (Status_Robot_x, Status_Robot_y, Status_Robot_RotationAngle). In this respect, it is understood that one or more of the aforementioned input signals can, for example, include odometry data that can be communicated by one or more motion sensors (for example, one or more of the aforementioned devices can include one or more motion sensors according to an embodiment of the disclosure) carried by the robot.

[0054] It is taken into account that, for example, the one or more casters and / or the one or more rectangular areas may be referred to as "items" according to one embodiment of the disclosure. It is further taken into account that a template may, for example, contain the following information / data: A) Size of the body (upper / lower limits) B) List of item groups, where each item group may include, for example, the following: I) Post type (swivel caster or rectangular post) Swivel caster: Ratio_of_wheel_offset_to_wheel_circumference (upper / lower limit) Wheel width to wheel circumference ratio (upper / lower limit) Wheel circumference (upper / lower limit) Rectangle: Item length (upper / lower limit) Item width (upper / lower limit) II) Position relative to the edge of the cart (upper / lower boundary on xy coordinates) III) Symmetry type (single item, left-right symmetry, front-back symmetry or four-way symmetry)

[0055] It is further taken into account that, for example, multiple templates can be configured for each deployment, and each template can be evaluated to determine a template associated with the best matching data according to an embodiment of the disclosure.

[0056] Regarding preprocessing: • Data relating to one or more wheel positions and / or one or more wheel orientations are determined, and / or • Deep data is processed

[0057] For example, the one or more input signals may include data that can be communicated by the one or more lidar sensors (for example, 2D lidar data) and / or data that can be communicated by the one or more depth sensors (for example, depth camera data), according to one embodiment of the disclosure.

[0058] In a specific example, 2D lidar data can be received and processed to determine the one or more wheel positions and / or orientations. The processing can be based on, for example, clustering-based processing (e.g., jump distance clustering or quadrant division of the vehicle area), and / or wheel-mount-based processing (e.g., random sampling consensus (RANSAC) processing, nonlinear least squares processing, and / or training a neural network to reset the wheel position from a subset of the 2D lidar data), and / or occlusion-based processing (e.g., determining which one or more sides of the mounted wheel are potentially occluded), or any combination thereof.For example, 2D lidar data can be processed in a manner of cluster-based processing, wheel-mount-based processing, and occlusion-characterization-based processing to determine the one or more wheel positions and / or orientations according to an embodiment of the disclosure.

[0059] In one embodiment, with regard to occlusion-characterization-based processing, it is taken into account that the lidar points next to the wheels can be used to determine whether a point is closer (i.e., an indication of occlusion) or farther away (i.e., an indication of non-occlusion). For the nearest detected corner of a wheel, it can be assumed that one or more of the following parameters associated with a wheel can be determined / detected: • Detected_wheel_x_corner, • Detected_Rad_y_Corner, • Detected wheel rotation angle, • Detected_wheel_observed_width, • Detected_wheel_observed_length, • Has lateral occlusion, • Has_coverage_in_longitudinal_direction

[0060] Furthermore, it can be assumed that, for example, a double covering of a wheel according to one embodiment of the disclosure is not possible.

[0061] In a specific example, depth camera data, which may include depth information, can be received and processed. For example, RANSAC template fitting can be used, and / or a neural network can receive and process bird's-eye view (BEV frame) data to output either a direct shape estimate or a segmented bitmap image that can be used, according to one embodiment of the disclosure. For example, it can be assumed that the nearest vertical surface of a cart can be characterized based on determining the position of the two nearest corners. • Has _left_side_observable • Detected_body_x_left • Detected_Body_y_left • Has_right_side_observable • Detected_body_x_right • Detected_Body_y_right

[0062] With regard to inference-based processing, it is considered that one or more statistical inference-based techniques (for example, nonlinear Bayes filters with sliding-window factor diagrams, such as extended Kalman filters, unscented Kalman filters, particle filters, etc., and variational Bayes-Kalman filters) may be used to derive an update of the overall status, which, according to one embodiment of the disclosure, may be associated with a cost function. The cost function may, for example, include one of the following parameters or any combination thereof: • Positional errors in the longitudinal direction with respect to a car body • Lateral positioning error between the corners of the car body, if detectable • Position of one or more wheels with reference to the position expected according to the condition assessment.

[0063] It is taken into account that in one embodiment, when a sliding window factor diagram is used, one or more key frames can be strategically selected, with some key frames being selected from the initial approach (for example, when the robot initially approaches a cart), even if the final alignment (for example, after the initial approach, when the robot performs a fine positioning alignment with respect to the cart in order to dock the cart) is underway.

[0064] In general, according to one embodiment of the disclosure, one or more input signals (which may include, for example, odometry data, 2D lidar data, and / or depth camera data) can be received (for example, by the first module 202) and processed, for example, by initialization, preprocessing, and / or inference-based processing (for example, by the second module 204) to generate one or more output signals that can be communicated (for example, via the third module 206). The one or more output signals may, for example, correspond to one or more control signals that can be used, for example, for the navigation of an AGV to facilitate cart transport (which may be linked, for example, to cart detection and / or cart pickup), according to one embodiment of the disclosure.

[0065] Accordingly, the present disclosure considers a sensor fusion approach that may be based, for example, on a combination of detecting one or more cart wheels and detecting the cart body, according to one embodiment of the disclosure. This can be helpful, for example, in cart docking (referred to, for example, as "car cart pickup"), in conjunction with carts equipped with steering wheels. According to one embodiment of the disclosure, it is taken into account that, in an example of a cart with steering wheels (which may be associated with arbitrary rotation about a vertical axis), ambiguities may arise when the cart position is reset from the detection of one or more (steering) wheels.

[0066] The above approach takes into account that at least one robust method (which, for example, addresses the aforementioned ambiguities) and / or one efficient method (which, for example, mitigates the need for pickup / detection aids) can be provided / facilitated for vehicle detection and / or pickup according to an embodiment of the disclosure.

[0067] The advantageous aspects of device 102 described above in the present disclosure can also be applied analogously to all aspects of a processing method described below in the present disclosure.

[0068] Likewise, all the advantageous aspects of the processing method described below can also be applied analogously to all aspects of the device 102 described above in the disclosure. It is understood that these remarks apply analogously to the system 100 discussed previously in the present disclosure.

[0069] Referring to Fig. 3 shows a processing method in connection with system 100 according to one embodiment of the disclosure. Processing method 300 may, for example, be suitable for facilitating cart transport according to one embodiment of the disclosure. Furthermore, processing method 300, or any section / part thereof, may, for example, possibly be carried out at any time according to one embodiment of the disclosure.

[0070] According to one embodiment of the disclosure, the processing method 300 may, for example, include initialization step 302, preprocessing step 304 and inference step 306 or any combination thereof.

[0071] In one embodiment, the processing method 300 can include an initialization step 302, a preprocessing step 304, and an inference step 306. In another embodiment, the processing method 300 can include an initialization step 302 and a preprocessing step 304. In yet another embodiment, the processing method 300 can include a preprocessing step 304 and an inference step 306. In yet another embodiment, the processing method 300 can include an inference step 306. In yet another additional embodiment, the processing method 300 can include an initialization step 302, a preprocessing step 304, and an inference step 306 (that is, an initialization step 302, a preprocessing step 304, or an inference step 306).In yet another additional embodiment, the processing method 300 can, for example, include an initialization step 302 and an inference step 306.

[0072] With reference to initialization step 302, according to one embodiment of the disclosure, one or more processing tasks can be performed in connection with the initialization of a standard status estimate (referred to as a "status estimate") in connection with at least one carriage. For example, the one or more devices 102 can be configured to perform the one or more processing tasks in connection with the initialization of a status estimate in connection with a carriage (for example, a status estimate of a carriage status associated with a carriage), as previously described with reference to Fig. 2 discussed, according to one embodiment of the disclosure.

[0073] With reference to preprocessing step 304, one or more processing tasks can be performed in conjunction with one or both of the following operations: • Determination of data regarding one or more wheel positions and / or one or more wheel orientations, and • Processing of in-depth data is performed, according to one embodiment of the disclosure.

[0074] For example, according to one embodiment of the disclosure, the one or more devices 102, as previously referred to, Fig. 2 discussed, be trained to perform one or more processing tasks in connection with one or both of the following operations: • Determination of data regarding one or more wheel positions and / or one or more wheel orientations • Processing of depth data

[0075] With reference to inference step 306, according to one embodiment of the disclosure, one or more processing tasks can be performed in conjunction with inference-based processing. For example, the inference-based processing can be combined with an inference of an update of the overall status. As mentioned previously, an update of the overall status can, for example, be combined with a cost function. For example, the one or more devices 102 can be configured to perform the one or more processing tasks in conjunction with the inference-based processing, as previously described with reference to Fig. 2 discussed, according to one embodiment of the disclosure.

[0076] It is understood that the steps (for example, the initialization step 302, the preprocessing step 304, and / or the inference step 306) need not be sequential steps per se. For example, several templates can be adjusted simultaneously to determine the best template in cases where, during initialization, insufficient data is available to distinguish a template according to one embodiment of the disclosure.

[0077] The present disclosure further includes a computer program (not shown) which may contain instructions which, when the program is executed by a computer (not shown), cause the computer to perform the initialization step 302, the preprocessing step 304 and / or the inference step 306 as discussed with reference to the processing procedure 300.

[0078] However, the present disclosure further includes a computer-readable storage medium (not shown) containing data stored therein that constitutes software executable by a computer (not shown), the software including instructions which, when executed by the computer, perform the initialization step 302, the preprocessing step 304 and / or the inference step 306, as discussed with reference to the processing method 300.

[0079] In light of the foregoing, it is understood that the present disclosure generally refers to a processing method 300, which may, for example, be suitable for facilitating wagon transport. In one embodiment, the processing method 300 may, for example, include an initialization step 302 and an inference step 306.

[0080] Referring to initialization step 302, a status estimate can be determined in conjunction with a cart status. The cart status can be associated with a cart. The cart can, for example, consist of a cart body and multiple wheels. The multiple wheels can be coupled to the cart body. In general, the wheels can be designed, for example, to facilitate the movement of the cart body.

[0081] Referring to inference step 306, an update of the overall status can be derived and linked to the cart. This update can be linked to a cost function. The cost function might include, for example, a parameter for longitudinal position errors with respect to the cart body, a parameter for lateral position errors between the corners of the cart body (if detectable), and a positioning parameter corresponding to the positioning of the wheels relative to the status estimate.

[0082] In one embodiment, the processing method 300 may, for example, further include a preprocessing step 304. With reference to the preprocessing step 304, the positions of the wheels and / or the orientations of the wheels may be determined, for example, based on the acquired data. Acquired data may, for example, be communicated by one or more devices 104 according to one embodiment of the disclosure. The one or more devices 104 may, for example, include one or more sensors. The one or more sensors may, for example, include at least one lidar sensor (for example, a 2D lidar sensor) and / or at least one depth sensor (for example, a depth camera). For example, acquired data may include lidar data (for example, 2D lidar data) and / or depth data (for example, camera depth data).In a specific example, data collected from at least one lidar sensor and / or at least one depth sensor can be communicated.

[0083] In one embodiment, captured data can be processed by clustering-based processing, wheel-mount-based processing, and occlusion-characterization-based processing, or any combination thereof (i.e., clustering-based processing and / or wheel-mount-based processing and / or occlusion-characterization-based processing; clustering-based processing, wheel-mount-based processing, and / or occlusion-characterization-based processing). For example, clustering-based processing can be based on jump-distance clustering or a division of the merchandise area into quadrants.In one example, wheel-attachment-based processing can be based on consensus random sampling (RANSAC) processing and / or non-linear least-squares processing and / or neural network training to reset the wheel position from a subset of the acquired data (that is, RANSAC and / or neural network training). In another example, occlusion-characterization-based processing can correspond to determining a potential occlusion of any of the wheels.

[0084] In one embodiment, inference step 306 may, for example, involve the execution of at least one processing task in conjunction with inference-based processing. The inference-based processing may, for example, be combined with at least one statistical inference approach-based technique. A statistical inference approach-based technique may, for example, be based on a sliding-window factor plot or a variational Bayes-Kalman filter. In an example using a sliding-window factor plot, multiple keyframes may be strategically selected, with some keyframes being chosen from an initial approach even while the final alignment is underway.

[0085] Furthermore, in light of the foregoing, it is understood that the present disclosure generally relates to a device 102 in conjunction with the processing method 300 according to one embodiment of the disclosure. The device 102 may, for example, be suitable for facilitating the transport of a cart. Moreover, the device 102 may, for example, correspond to an electronic module 200a, as previously described with reference to Fig. 2 discussed, according to one embodiment of the disclosure.

[0086] The device 102 can, for example, include a first module 202, a second module 204 and a third module 206, according to one embodiment of the disclosure.

[0087] In one embodiment, the first module 202 (which corresponds, for example, to a receiver) can be configured, for example, to receive at least one input signal (which can be communicated, for example, by at least one device 104). The second module 204 (which corresponds, for example, to a processor) can be configured, for example, to process the one or more input signals according to the processing method 300 as discussed above, in order to generate at least one output signal. The third module 206 (which corresponds, for example, to a transmitter) can be configured, for example, to communicate the one or more output signals. The one or more output signals can be communicated, for example, to a robot (for example, an AGV) and / or one or more devices 104. The one or more output signals can be used, for example, to facilitate cart transport (for example, for cart detection and / or cart retrieval).

[0088] Accordingly, the present disclosure considers a sensor fusion approach that may be based, for example, on a combination of detecting one or more cart wheels and detecting the cart body, according to one embodiment of the disclosure. This can be helpful, for example, in cart docking (referred to, for example, as "car cart pickup"), in conjunction with carts equipped with steering wheels. According to one embodiment of the disclosure, it is taken into account that, in an example of a cart with steering wheels (which may be associated with arbitrary rotation about a vertical axis), ambiguities may arise when the cart position is reset from the detection of one or more (steering) wheels.

[0089] The above approach takes into account that at least one robust method (which, for example, addresses the aforementioned ambiguities) and / or one efficient method (which, for example, mitigates the need for pickup / detection aids) can be provided / facilitated for vehicle detection and / or pickup according to an embodiment of the disclosure.

[0090] It is understood that the embodiments described above may be combined in any way where appropriate (for example, one or more embodiments as described in the section “Detailed Description” may be combined with one or more embodiments as described in the section “Abstract Description of the Invention”).

[0091] Furthermore, the person skilled in the art should note that variations and combinations of the embodiments described above, which are not alternatives or replacements, can be combined to form yet other embodiments.

[0092] In one example, the communication network 106 can be omitted. Communication (that is, between the one or more devices 102 and the one or more devices 104) can be achieved by direct coupling. Such direct coupling can be achieved by wired coupling and / or wireless coupling. For example, the one or more devices 102 and the one or more devices 104 can be carried by an AGV, and the one or more devices 102 and the one or more devices 104 can be directly coupled according to one embodiment of the disclosure.

[0093] In another example, estimates of the wheel's rotation angles can be used to determine if a cart has been disturbed, allowing the robot to abort the docking maneuver (i.e., the final alignment), move away from the cart, and repeat the initial approach.

[0094] In the aforementioned manner, various embodiments of the disclosure for addressing at least one of the aforementioned disadvantages are described. Such embodiments are to be encompassed by the following claims and are not limited to specific shapes or arrangements of parts so described, and it will be obvious to the person skilled in the art in view of this disclosure that numerous changes and / or modifications can be made, which are also to be encompassed by the following claims. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited non-patent literature

[0000] Bostelman, Roger, Roger Bostelman and Tsai Hong. Review of Research for Docking Automatic Guided Vehicles and Mobile Robots. US Department of Commerce, National Institute of Standards and Technology, 2016

[0003]

Claims

[1] Processing method (300) suitable for facilitating wagon transport, wherein the processing method (300) comprises the following: an initialization step (302) wherein a status estimate is determined in connection with a wagon status associated with a wagon, the wagon having a wagon body and multiple wheels associated with the wagon body, and an inference step (306) wherein an update of the overall status that can be associated with the wagon can be derived, wherein the update of the overall status is associated with a cost function, where the cost function has the following characteristics: a parameter for positional errors in the longitudinal direction with respect to the car body, a parameter for lateral positional errors between the corners of the car body, if detectable, and Positioning parameters that correspond to the positioning of the wheels in relation to the status estimate. [2] Processing method (300) according to claim 1, further comprising a preprocessing step (304) wherein positions of the wheels and / or orientations of the wheels are determined based on acquired data. [3] Processing method (300) according to one of the preceding claims, further comprising a preprocessing step (304) wherein depth data are determined based on acquired data. [4] Processing method (300) according to one of the preceding claims, wherein acquired data from at least one light detection and distance measurement (Lidar) sensor and / or at least one depth sensor can be communicated. [5] Processing method (300) according to any of the preceding claims, wherein acquired data are processed in a manner based on clustering-based processing and / or wheel attachment-based processing and / or occlusion characterization-based processing. [6] Processing method (300) according to one of the preceding claims, wherein the clustering-based processing is based on jump distance clustering or division of the carriage area into quadrants. [7] Processing method (300) according to any of the preceding claims, wherein the wheel attachment-based processing is based on consensus random sampling (RANSAC) processing, non-linear least squares processing, and / or training of a neural network to reset the wheel position from a subset of the acquired data. [8] Processing method (300) according to one of the preceding claims, wherein the cover characterization-based processing corresponds to the determination of a possible cover of one of the wheels. [9] Processing method (300) according to any of the preceding claims, wherein the inference step (306) includes the execution of at least one processing task in conjunction with inference-based processing. [10] Processing method (300) according to one of the preceding claims, wherein inference-based processing is combined with at least one statistical inference approach-based technique. [11] Processing method (300) according to one of the preceding claims, wherein a statistical inference approach-based technique is based on a sliding window factor diagram or a variational Bayes-Kalman filter, [12] Processing method (300) according to one of the preceding claims, wherein, where sliding window factor diagram is used, several key frames are strategically selected, with some key frames being selected from an initial approach even when the final alignment is in progress. [13] Computer program with instructions which, when the program is executed by a computer, cause the computer to perform the initialization step (302) and / or the preprocessing step (304) and / or the inference step (306) according to the processing method (300) of any of the preceding claims. [14] Computer-readable storage medium on which data is stored that represents computer-executable software, wherein the software includes instructions which, when executed by the computer, perform the initialization step (302) and / or the preprocessing step (304) and / or the inference step (306) according to the processing method (300) of any one of claims 1 to 12. [15] Device (102) comprising the following: a first module (202) that can be trained to receive at least one input signal; a second module (204) which can be configured to process the input signal according to the processing method (300) according to any one of claims 1 to 12, in order to generate at least one output signal; and a third module (206) that can be trained to communicate at least one output signal which can be used to facilitate wagon transport.

Images