Method, device, equipment, storage medium and program product for robot to identify trailer posture

Abstract

The application relates to a method, device, equipment, storage medium and program product for robot to identify a trailer posture. The method comprises the following steps: acquiring prior data of a trailer and radar point cloud collected by a robot; filtering the radar point cloud based on a detection area of the trailer to obtain first target radar point cloud; clustering the first target radar point cloud to obtain first clustering point cloud, and searching for a clustering point pair matched with the prior data in the first clustering point cloud to obtain a first candidate clustering point pair; determining a first angle of the trailer according to the first candidate clustering point pair; and identifying a posture of the trailer according to the prior data and the first angle. The method can enable the robot to accurately and quickly identify the posture of the trailer.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robots, in particular to a method and device for recognizing a posture of a trailer by a robot, equipment, a storage medium and a program product. BACKGROUND

[0002] With the development of automation and artificial intelligence, more and more robots are applied to production and life. At present, many factories have popularized the use of robots to tow material racks or mobile pallets to transport materials or goods. In the existing towing system, only the obstacle avoidance of the robot is considered, but the obstacle avoidance of the towed trailer is not considered. SUMMARY

[0003] Therefore, it is necessary to provide a method and device for recognizing a posture of a trailer by a robot, equipment, a storage medium and a program product to enable the robot to accurately and quickly recognize the posture modeling of the trailer for obstacle avoidance.

[0004] In a first aspect, the present application provides a method for recognizing a posture of a trailer by a robot, the method comprising:

[0005] obtaining prior data of the trailer and radar point clouds collected by the robot, wherein the prior data comprises a size of the trailer, a preset constraint angle between the trailer and a towing device, and a length of the towing device;

[0006] filtering the radar point clouds based on a detection area of the trailer to obtain first target radar point clouds;

[0007] performing clustering on the first target radar point clouds to obtain first clustered point clouds, and finding a clustered point pair matching the prior data in the first clustered point clouds to obtain a first candidate clustered point pair;

[0008] determining a first angle of the trailer according to the first candidate clustered point pair;

[0009] recognizing a posture of the trailer according to the prior data and the first angle.

[0010] In a second aspect, the present application further provides a device for recognizing a posture of a trailer by a robot, the device comprising:

[0011] an obtaining module configured to obtain prior data of the trailer and radar point clouds collected by the robot, wherein the prior data comprises a size of the trailer, a preset constraint angle between the trailer and a towing device, and a length of the towing device;

[0012] a filtering module configured to filter the radar point clouds based on a detection area of the trailer to obtain first target radar point clouds;

[0013] a clustering module, configured to cluster the first target radar point cloud to obtain a first clustered point cloud;

[0014] a searching module, configured to search for a clustered point pair matching the prior data from the first clustered point cloud to obtain a first candidate clustered point pair;

[0015] a determining module, configured to determine a first angle of the trailer according to the first candidate clustered point pair;

[0016] an identifying module, configured to identify a pose of the trailer according to the prior data and the first angle.

[0017] In a third aspect, the present application provides a computer device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of the method for identifying a pose of a trailer when executing the computer program.

[0018] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the method for identifying a pose of a trailer when executed by a processor.

[0019] In a fifth aspect, the present application provides a computer program product, which comprises a computer program, and the computer program implements the steps of the method for identifying a pose of a trailer when executed by a processor.

[0020] The above method, device, computer device, storage medium and computer program product for identifying a pose of a trailer can obtain prior data of a trailer and radar point cloud collected by a robot, filter the radar point cloud based on a detection area of the trailer to obtain a first target radar point cloud, thereby reducing the number of radar point clouds and effectively reducing the calculation amount, which is conducive to quickly detecting the pose of the trailer; the first target radar point cloud is clustered to obtain a first clustered point cloud, and a clustered point pair matching the prior data is searched from the first clustered point cloud to obtain a first candidate clustered point pair; the first angle of the trailer is determined according to the first candidate clustered point pair; and the pose of the trailer can be quickly and accurately identified according to the prior data and the first angle, and modeling can be performed by using the pose, which can effectively avoid the trailer from colliding with an obstacle during travel. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 FIG. 1 is a schematic diagram of an application environment of the method for identifying a pose of a trailer by a robot in an embodiment;

[0022] Figure 2 FIG. 2 is a schematic diagram of a flow of the method for identifying a pose of a trailer by a robot in an embodiment.

[0023] Figure 3 A schematic diagram of a system for towing a trailer by a robot in an embodiment;

[0024] Figure 4 A schematic diagram of a detection area in an embodiment;

[0025] Figure 5 A block diagram of an apparatus for a robot to recognize a posture of a trailer in an embodiment;

[0026] Figure 6 An internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0027] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0028] It should be noted that in the description of the present application, the terms "first, second, third and fourth" are only used to distinguish similar objects, and do not represent a specific order of the objects. It can be understood that "first, second, third and fourth" can be interchanged in a specific order or sequence as allowed, so that the described embodiments of the present application can be implemented in an order other than that illustrated or described.

[0029] An embodiment of the present application provides a method for a robot to recognize a posture of a trailer, which can be applied to an application scenario as shown in Figure 1 The robot 110 can be configured as any intelligent device with the functions of obstacle sensing and autonomous navigation, such as an autonomous guided vehicle (AGV / AMR), a cleaning robot, a delivery robot, a welcome robot, etc.; the trailer 130 can be configured as various racks, carts, trolleys, etc. with casters and accommodating spaces. The robot 110 is connected to the trailer 130 through a towing device 120 such as a towing rod, a towing hook, a towing chain, a towing rope, etc. In the illustrated embodiment, the robot 110 is provided with a laser radar 111 at each of the right front corner and the left rear corner, and the robot 110 obtains radar point cloud data through the laser radar 111 for environmental sensing.

[0030] As shown in Figure 2 In an embodiment, a method for a robot to recognize a posture of a trailer is provided, which can be executed by the robot in Figure 1 The method comprises the following steps:

[0031] S202, acquire prior data of the trailer and radar point cloud collected by the robot. The prior data includes the size of the trailer, the preset constraint angle between the trailer and the towing device, and the length of the towing device.

[0032] S204: Filter the radar point cloud based on the detection area of ​​the trailer to obtain the first target radar point cloud.

[0033] S206, cluster the first target radar point cloud to obtain a first clustered point cloud, and find the cluster point pairs that match the prior data in the first clustered point cloud to obtain the first candidate cluster point pairs. The first clustered point cloud can be multiple clusters (which can be called sub-clustered point clouds) resulting from the clustering processing of the first target radar point cloud. A cluster point can be a sub-clustered point cloud in the first clustered point cloud; correspondingly, a cluster point pair can be a combination of two sub-clustered point clouds in the first clustered point cloud.

[0034] S208, determine the first angle of the trailer based on the first candidate cluster point pair, wherein the first angle can be the angle of the trailer relative to the robot, such as the angle of the trailer relative to the robot.

[0035] S210 identifies the trailer's posture based on prior data and the first angle.

[0036] In the above embodiments, prior data of the trailer and radar point clouds collected by the robot are acquired; the radar point clouds are filtered based on the detection area of ​​the trailer to obtain a first target radar point cloud, thereby reducing the number of radar point clouds, effectively reducing the amount of computation, and facilitating the rapid detection of the trailer's attitude; in addition, the first target radar point cloud is clustered to obtain a first cluster point cloud, and cluster point pairs that match the prior data are found in the first cluster point cloud to obtain a first candidate cluster point pair; the first angle of the trailer is determined based on the first candidate cluster point pair; thus, the attitude of the trailer can be quickly and accurately identified based on the prior data and the first angle, and the attitude can be used for modeling, which can effectively prevent the trailer from encountering obstacles during travel.

[0037] like Figure 3 As shown, the dimensions of the trailer are used to characterize the size of the trailer, which includes the length (L) and width (W) of the trailer; the preset constraint angle (θ) between the trailer and the towing device is used to characterize the deviation angle of the trailer relative to the robot, that is, the angle between the trailer and the towing device. This angle can be preset to 80° to 100° or 85° to 95°, and the specific range can be adjusted according to the actual constraint requirements.

[0038] In one embodiment, the robot acquires the inputted prior data for the trailer in response to the data input operation. For example, the robot displays an input page on the screen, on which the user can input at least one of the size of the trailer, the preset constraint angle between the trailer and the towing device, or the length of the towing device.

[0039] In one embodiment, the robot emits a laser beam through the installed laser radar, which is reflected when encountering an "obstacle", so that the reflected laser beam can be collected to obtain a radar point cloud. The obstacle can include the trailer, and can also include other objects that encounter the laser beam, such as environmental objects and pedestrians.

[0040] Preferably, a reflector or a reflective sticker can be arranged on the two side edges of the front side of the trailer, so that the robot can identify the distance between the two reflectors or reflective stickers through the radar device to identify the width of the trailer, and then determine the position of the trailer in combination with the length data of the trailer.

[0041] For example, the trailer can swing left and right within a certain angle range during the process of traveling under the traction of the robot. The size of the trailer and the preset constraint angle can be used to obtain the area (i.e., the detection area) where the reflector can be located, such as a C-shaped area. Figure 4 In addition, the size of the trailer and the preset constraint angle can also be used to obtain the detection area where the trailer can be located.

[0042] In one embodiment, the first target radar point cloud in S204 can be the radar point cloud located in the detection area.

[0043] In one embodiment, the detection area can be determined according to the size of the trailer and the preset constraint angle.

[0044] It can be understood that, during the process of traveling under the traction of the robot, the trailer can deviate from the robot due to the obstacle avoidance actions of the robot such as turning and braking; or the trailer can deviate left and right from the rear of the robot due to uneven ground, so that the trailer and the robot form a certain angle, and thus the trailer can not be always located at the rear of the robot, but can be located at the rear or the oblique rear of the robot. Therefore, in order to reduce the calculation amount and ensure the realization of the posture detection of the trailer, a detection area can be determined at the rear (including the rear and the oblique rear) of the robot, which is a C-shaped area at the rear of the robot, and can be used to represent a set of possible positions of the trailer, so as to perform the posture detection by using the radar point cloud located in the detection area.

[0045] In one embodiment, after obtaining the first target radar point cloud in S204, radar points in the first target radar point cloud that do not satisfy the reflection intensity threshold can also be filtered out to obtain a second target radar point cloud; the second target radar point cloud is clustered to obtain a second clustered point cloud; and a clustered point pair matching the prior data is searched for from the second clustered point cloud.

[0046] Further, if a clustered point pair matching the prior data is not searched for from the second clustered point cloud, S206 is continued. It can be understood that, when the robot collects radar point cloud data by the laser radar, the reflection intensity corresponding to some angles can be too low or the information can be wrong, which can cause a clustered point pair matching the prior data to be unable to be searched for from the second target radar point cloud after the intensity filtering and clustering.

[0047] In S206, the first clustered point cloud includes a "point type clustered point cloud" and a "line type clustered point cloud". In one embodiment, S206 can specifically include:

[0048] Each clustered point in the point type clustered point cloud and each edge clustered point in the line type clustered point cloud are combined two by two to obtain a plurality of clustered point pairs.

[0049] A third angle between a line of each clustered point pair in the plurality of clustered point pairs and a virtual traction device is determined, wherein the virtual traction device is a virtual connection line between a midpoint of the line and a traction point (m) of the robot, i.e., a connection line between the midpoint of the line of the clustered point pair and the traction point (m) of the robot.

[0050] According to a matching degree between the third angle and a preset constraint angle, a matching degree between a length of the virtual traction device and a length of the traction device, a matching degree between a length of the line and a trailer width, and a matching degree between the first angle and a first angle calculated in a previous frame, a first candidate clustered point pair is selected from the first clustered point cloud.

[0051] In the line type clustered point cloud, the edge clustered point includes a front end point (i.e., a front end clustered point) and a rear end point (i.e., a rear end clustered point). When the clustered points are combined, each clustered point in the point type clustered point cloud can be combined, a clustered point in the point type clustered point cloud and the front end point can be combined, a clustered point in the point type clustered point cloud and the rear end point can be combined, and the front end point and the rear end point in the point type clustered point cloud can be combined.

[0052] For example, if 20 clustered point pairs are obtained by combining the clustered points in the first clustered point cloud, the clustered point pairs that satisfy the following conditions are selected as candidate clustered point pairs from the 20 clustered point pairs:

[0053] a) whether the angle of the virtual towing device between the midpoint of the line and the robot towing point matches a preset constraint angle;

[0054] b) whether the length of the virtual towing device between the midpoint of the line and the robot towing point is close to the length of the towing device in the prior data;

[0055] c) whether the length of the line is close to the width of the trailer in the prior data;

[0056] d) whether the angle of the virtual towing device and the line is close to the angle detected in the previous frame.

[0057] In an embodiment, after the candidate clustering point pair is selected, the first angle of the trailer is determined according to the first candidate clustering point pair, and further screening is performed on the basis of the score obtained according to the above conditions. Specifically, the robot determines a fifth score according to the matching degree between the third angle and the preset constraint angle, determines a sixth score according to the matching degree between the length of the virtual towing device and the length of the towing device, determines a seventh score according to the matching degree between the length of the line and the width of the trailer, and determines an eighth score according to the matching degree between the third angle and the third angle calculated in the previous frame. The fifth score, the sixth score, the seventh score and the eighth score are weighted and summed to obtain a second comprehensive score. The first candidate clustering point cloud is screened based on the second comprehensive score to obtain a screened first candidate clustering point pair. For example, the first candidate clustering point pair with the largest second comprehensive score is taken as the screened first candidate clustering point pair, and the screened first candidate clustering point pair is updated to the first candidate clustering point, and then the first angle of the trailer is determined according to the updated first candidate clustering point.

[0058] For the point type clustering point cloud and the line type clustering point cloud, the robot can determine whether each sub-clustering point cloud in the first clustering point cloud is a point type clustering point cloud or a line type clustering point cloud by the following method: the robot calculates the circumscribed rectangle of each sub-clustering point cloud in the first clustering point cloud; the sub-clustering point cloud with a long side of the circumscribed rectangle less than a preset threshold is determined as a point type clustering point cloud; and the sub-clustering point cloud with a long side of the circumscribed rectangle greater than or equal to the preset threshold is determined as a line type clustering point cloud.

[0059] In an embodiment, on the contrary, if a clustering point pair matching the prior data can be found from the second clustering point cloud, the matching clustering point pair is taken as the second candidate clustering point pair; the second angle of the trailer is determined according to the second candidate clustering point pair; and the posture of the trailer is identified according to the prior data and the second angle.

[0060] In one embodiment, after the robot selects the candidate cluster point pair, before determining the second angle of the trailer according to the second candidate cluster point pair, the robot further screens in combination with the score obtained according to the above conditions, specifically: the robot determines a first score according to the matching degree between the first angle and the preset constraint angle, determines a second score according to the matching degree between the length of the virtual traction device and the length of the traction device, determines a third score according to the matching degree between the length of the connecting line and the width of the size, and determines a fourth score according to the matching degree between the first angle and the first angle obtained in the previous frame calculation; the first score, the second score, the third score and the fourth score are weighted and summed to obtain a first comprehensive score; the second candidate cluster point pair is screened based on the first comprehensive score to obtain a screened second candidate cluster point. For example, the first comprehensive score of the second candidate cluster point pair is taken as the screened second candidate cluster point pair, and the screened second candidate cluster point pair is updated as the second candidate cluster point, and then the second angle of the trailer is determined according to the updated second candidate cluster point pair.

[0061] In the above embodiment, after the first target radar point cloud is obtained by filtering the radar point cloud based on the detection area, the first target radar point cloud is further filtered according to the reflection intensity, so that the point cloud that does not meet the intensity condition is screened out, so that the speed of posture recognition can be accelerated; in addition, the second target radar point cloud is obtained by clustering, the matched cluster point pair is selected from the second cluster point cloud as the second candidate cluster point pair, and the second angle of the trailer is determined according to the second candidate cluster point pair; so that the posture of the trailer can be quickly and accurately recognized according to the prior data and the second angle, and then the posture is used for modeling, which can effectively avoid the trailer from colliding with obstacles during travel.

[0062] In combination with a specific example, the technical solutions of the present application are described as follows:

[0063] S1, input the length (L) and width (W) of the trailer, the preset constraint angle (θ) between the trailer and the traction device, and the length (D) of the traction device;

[0064] S2, calculate the mask according to the preset constraint angle and the length (L) and width (W) of the trailer, and select the radar point cloud on the mask;

[0065] Wherein, the mask mentioned above refers to the detection area, which is a set of all possible positions of the trailer at different angles.

[0066] S3, filter out the radar point cloud outside the mask to obtain the first cluster point cloud, that is, the radar point cloud located in the mask, denoted as mask_filter_points;

[0067] S4, filter out the radar point cloud in mask_filter_points whose radar reflection intensity is below 235, to obtain the second clustering point cloud, i.e. the radar point cloud whose radar reflection intensity is above 235, denoted as inte_filter_points;

[0068] S5, first cluster inte_filter_points to obtain a clustering point cloud set Φ;

[0069] S6, judge any two clustering points in the radar clustering point cloud in the clustering point cloud set Φ to find out a candidate clustering point pair φ, wherein the judgment is based on the following conditions:

[0070] S61) whether the angle between the virtual towing device between the midpoint of the connecting line and the robot towing point and the connecting line matches the preset constraint angle θ;

[0071] S62) whether the length of the virtual towing device between the midpoint of the connecting line and the robot towing point is close to the length of the towing device in the prior data;

[0072] S63) whether the length of the connecting line is close to the width of the trailer in the prior data;

[0073] S64) whether the angle of the virtual towing device and the connecting line is close to the angle detected in the last frame.

[0074] S7, weight the matching degree of the conditions in the candidate clustering point pair φ in the last step to obtain a score, and select the clustering point pair with the highest score as the best candidate clustering point pair;

[0075] S8, if no candidate clustering point pair satisfying the conditions is found in inte_filter_points, use mask_filter_points to find;

[0076] S9, cluster mask_filter_points to obtain a clustering point cloud set Ω;

[0077] S10, calculate the minimum bounding rectangle of each radar clustering point cloud in the clustering point cloud set Ω, and judge the point type clustering point cloud and the line type clustering point cloud according to the side length of the minimum bounding rectangle, as follows:

[0078] 1) when the long side of the minimum bounding rectangle is greater than a preset threshold, it is considered that this clustering point cloud is a line type clustering point cloud, and two edge points of this cluster are found according to the angle of each point in the line type clustering point cloud, and the edge points are denoted as "front end point" and "rear end point" according to the angle;

[0079] 2) when the long side of the minimum bounding rectangle is less than a preset threshold, it is considered that this clustering point cloud is a point type clustering point cloud.

[0080] S11, rules of point type cluster point cloud-point type cluster point cloud, front end point-back end point, point type cluster point cloud-back end point, and front end point-point type cluster point cloud are combined two by two, and the following conditions are used to find candidate cluster point pairs ω:

[0081] S111) whether the angle of the virtual traction device between the midpoint of the connecting line and the robot traction point matches the preset constraint angle θ;

[0082] S112) whether the length of the virtual traction device between the midpoint of the connecting line and the robot traction point is close to the length D of the traction device in the prior data;

[0083] S113) whether the length of the connecting line is close to the width W of the trailer in the prior data;

[0084] S114) whether the angle of the virtual traction device and the connecting line is close to the angle detected in the last frame.

[0085] S12, condition matching degree of candidate cluster point pairs is weighted, and the highest score is selected as the best candidate cluster point pair;

[0086] S13, the angle of the trailer is obtained according to the best candidate cluster point, and the overall pose of the trailer is calculated according to the size and angle of the trailer.

[0087] It should be understood that although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise stated herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps.

[0088] Based on the same inventive concept, an embodiment of the present application also provides a device for implementing the method of robot recognizing trailer pose described above. The implementation scheme of the device for solving the problem is similar to the implementation scheme described in the above method, so the specific limitations in one or more device embodiments for robot recognizing trailer pose provided below can refer to the limitations of the method of robot recognizing trailer pose described above, which will not be repeated here.

[0089] In one embodiment, as Figure 5As shown, a device for recognizing a posture of a trailer by a robot is provided, comprising: an acquisition module 502, a filtering module 504, a clustering module 506, a searching module 508, a determining module 510, and a recognizing module 512, wherein:

[0090] The acquisition module 502 is configured to acquire prior data of the trailer and radar point clouds collected by the robot, the prior data comprising a size of the trailer, a preset constraint angle between the trailer and a towing device, and a length of the towing device.

[0091] The filtering module 504 is configured to filter the radar point clouds based on a detection area of the trailer to obtain first target radar point clouds.

[0092] The clustering module 506 is configured to cluster the first target radar point clouds to obtain first clustered point clouds.

[0093] The searching module 508 is configured to search for clustered point pairs matching the prior data from the first clustered point clouds to obtain first candidate clustered point pairs.

[0094] The determining module 510 is configured to determine a first angle of the trailer according to the first candidate clustered point pairs.

[0095] The recognizing module 512 is configured to recognize a posture of the trailer according to the prior data and the first angle.

[0096] In one of the embodiments, the filtering module is further configured to determine a detection area of the trailer according to the size of the trailer and the preset constraint angle, and filter out radar point clouds located outside the detection area to obtain the first target radar point clouds.

[0097] In one of the embodiments, the device further comprises:

[0098] The filtering module is configured to filter out radar point clouds in the first target radar point clouds whose reflection intensity does not satisfy a reflection intensity threshold to obtain second target radar point clouds.

[0099] The clustering module is further configured to cluster the second target radar point clouds to obtain second clustered point clouds.

[0100] The searching module is further configured to search for clustered point pairs matching the prior data from the second clustered point clouds.

[0101] The clustering module is further configured to, if no clustered point pairs matching the prior data are found from the second clustered point clouds, continue to cluster the first target radar point clouds to obtain first clustered point clouds.

[0102] The searching module is further configured to search for clustered point pairs matching the prior data from the first clustered point clouds to obtain first candidate clustered point pairs.

[0103] In one of the embodiments, the device further comprises:

[0104] The determining module is further configured to determine a first angle between a line connecting each cluster point pair in the second cluster point cloud and a virtual towing device, the virtual towing device being a virtual connection line between a midpoint of the line and a towing point of the robot;

[0105] The searching module is further configured to search, from the second cluster point cloud, a cluster point pair matching the prior data based on a matching degree between the first angle and a preset constraint angle, a matching degree between a length of the virtual towing device and a length of the towing device, a matching degree between a length of the line and a width of the size of the trailer, and a matching degree between the first angle and a first angle calculated in a previous frame;

[0106] The determining module is further configured to, if the cluster point pair matching the prior data is found from the second cluster point cloud, take the matching cluster point pair as a second candidate cluster point pair, and determine a second angle of the trailer according to the second candidate cluster point pair;

[0107] The identifying module is further configured to identify the posture of the trailer according to the prior data and the second angle.

[0108] In one of the embodiments, the searching module is further configured to determine a first score according to the matching degree between the first angle and the preset constraint angle, determine a second score according to the matching degree between the length of the virtual towing device and the length of the towing device, determine a third score according to the matching degree between the length of the line and the width of the size, and determine a fourth score according to the matching degree between the first angle and the first angle calculated in the previous frame; and perform weighted summation on the first score, the second score, the third score and the fourth score to obtain a first comprehensive score; and perform screening on the second candidate cluster point pair based on the first comprehensive score to obtain a screened second candidate cluster point.

[0109] In the above embodiments, after the first target radar point cloud is obtained by filtering the radar point cloud based on the detection area, the first target radar point cloud is further filtered according to the reflection intensity, so as to screen out the point cloud that does not meet the condition, thereby accelerating the posture recognition speed; and the second cluster point cloud is obtained by clustering the obtained second target radar point cloud, the matching cluster point pair is selected from the second cluster point cloud as the second candidate cluster point pair, and the second angle of the trailer is determined according to the second candidate cluster point pair; thereby the posture of the trailer can be quickly and accurately identified according to the prior data and the second angle, and modeling is performed by using the posture, so as to effectively avoid the trailer from colliding with obstacles during the travel.

[0110] In one of the embodiments, the first cluster point cloud comprises a point type cluster point cloud and a line type cluster point cloud.

[0111] The searching module is further configured to: combine each cluster point in the point-type cluster point cloud and each edge cluster point in the line-type cluster point cloud in pairs to obtain a plurality of cluster point pairs; determine a third angle between a connecting line of each cluster point pair in the plurality of cluster point pairs and a virtual traction device, the virtual traction device being a virtual connection line between a midpoint of the connecting line and a traction point of the robot; and select a first candidate cluster point pair from the first cluster point cloud according to a matching degree between the third angle and a preset constraint angle, a matching degree between a length of the virtual traction device and a length of the traction device, a matching degree between a length of the connecting line and a width in a size, and a matching degree between the third angle and a third angle obtained in a previous frame.

[0112] In one of the embodiments, the searching module is further configured to: determine a fifth score according to the matching degree between the third angle and the preset constraint angle, determine a sixth score according to the matching degree between the length of the virtual traction device and the length of the traction device, determine a seventh score according to the matching degree between the length of the connecting line and the width in the size, and determine an eighth score according to the matching degree between the third angle and the third angle obtained in the previous frame; obtain a second comprehensive score by weighted summation of the fifth score, the sixth score, the seventh score, and the eighth score; and filter the first candidate cluster point cloud based on the second comprehensive score to obtain a filtered first candidate cluster point pair.

[0113] In one of the embodiments, the device further includes:

[0114] The calculating module is configured to calculate a circumscribed rectangle for each sub-cluster point cloud in the first cluster point cloud.

[0115] The determining module is configured to determine a sub-cluster point cloud with a long side of the circumscribed rectangle being less than a preset threshold as a point-type cluster point cloud, and determine a sub-cluster point cloud with a long side of the circumscribed rectangle being greater than or equal to the preset threshold as a line-type cluster point cloud.

[0116] The above-mentioned modules in the device for recognizing a trailer posture of a robot can be realized by software, hardware, or a combination thereof, in whole or in part. The above-mentioned modules can be embedded in or independent of a processor in a computer device in a hardware form, or stored in a memory in the computer device in a software form, so as to be called and executed by a processor to perform operations corresponding to the above-mentioned modules.

[0117] In one of the embodiments, a computer device is provided, which can be a robot, and an internal structure diagram of the computer device can be as shown in Figure 6As shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with the external terminal in a wired or wireless manner. The wireless manner can be realized through WIFI, mobile cellular network, NFL (near field communication) or other technologies. The computer program is executed by the processor to realize a method for identifying the posture of a trailer by a robot. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device. It can also be an external keyboard, touchpad or mouse, etc.

[0118] Those skilled in the art can understand that, Figure 6 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0119] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method for identifying the posture of a trailer by a robot.

[0120] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method for identifying the posture of a trailer by a robot.

[0121] In one embodiment, a computer program product is provided, including a computer program, and the computer program is executed by a processor to implement the steps of the method for identifying the posture of a trailer by a robot.

[0122] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0123] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0124] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method of a robot identifying a trailer attitude, characterized by, The method comprises: acquiring prior data of a trailer and radar point cloud collected by a robot, the prior data comprising a size of the trailer, a preset constraint angle between the trailer and a towing device, and a length of the towing device; filtering the radar point cloud based on a detection area of the trailer to obtain a first target radar point cloud; performing clustering on the first target radar point cloud to obtain a first clustered point cloud, and searching for a clustered point pair matching the prior data in the first clustered point cloud to obtain a first candidate clustered point pair; the first clustered point cloud comprises a point type clustered point cloud and a line type clustered point cloud; determining a first angle of the trailer according to the first candidate clustered point pair; identifying a posture of the trailer according to the prior data and the first angle; wherein the searching for the clustered point pair matching the prior data in the first clustered point cloud to obtain the first candidate clustered point pair comprises: combining each clustered point in the point type clustered point cloud with each edge clustered point in the line type clustered point cloud in pairs to obtain a plurality of clustered point pairs; determining a third angle between a connecting line of each clustered point pair in the plurality of clustered point pairs and a virtual towing device, the virtual towing device being a virtual connecting line between a midpoint of the connecting line and a towing point of the robot; and selecting a first candidate clustered point pair from the first clustered point cloud according to a matching degree between the third angle and the preset constraint angle, a matching degree between a length of the virtual towing device and a length of the towing device, a matching degree between a length of the connecting line and a width in the size, and a matching degree between the third angle and a third angle obtained in a previous frame.

2. The method of claim 1, wherein, The filtering of the radar point cloud based on the detection area of the trailer to obtain the first target radar point cloud comprises: determining a detection area of the trailer according to the size and the preset constraint angle; filtering out radar point clouds located outside the detection area to obtain the first target radar point cloud.

3. The method according to claim 1 or 2, characterized in that, The method further comprises: filtering out radar point clouds in the first target radar point cloud that do not satisfy a reflection intensity threshold to obtain a second target radar point cloud; performing clustering on the second target radar point cloud to obtain a second clustered point cloud, and searching for a clustered point pair matching the prior data from the second clustered point cloud; if no clustered point pair matching the prior data is found from the second clustered point cloud, continuing to perform the steps of performing clustering on the first target radar point cloud to obtain a first clustered point cloud, and searching for a clustered point pair matching the prior data in the first clustered point cloud to obtain a first candidate clustered point pair.

4. The method of claim 3, wherein, The searching for a clustered point pair matching the prior data from the second clustered point cloud comprises: determining a first angle between a connecting line of each clustered point pair in the second clustered point cloud and a virtual towing device, the virtual towing device being a virtual connecting line between a midpoint of the connecting line and a towing point of the robot; The matching cluster point pairs matching the prior data are searched from the second cluster point cloud based on matching degrees between the first angles between the lines of the cluster point pairs and the virtual towing device and the preset constraint angle, matching degrees between lengths of the virtual towing device and the length of the towing device, matching degrees between lengths of the lines and the width in the size, and matching degrees between the first angles between the lines of the cluster point pairs and the first angles calculated in the last frame.

5. The method of claim 4, wherein, The method further comprises: If the matching cluster point pairs matching the prior data are searched from the second cluster point cloud, the matching cluster point pairs are taken as second candidate cluster point pairs; a second angle of the trailer is determined according to the second candidate cluster point pairs; a posture of the trailer is identified according to the prior data and the second angle.

6. The method of claim 5, wherein, Before the second angle of the trailer is determined according to the second candidate cluster point pairs, the method further comprises: a first score is determined according to the matching degrees between the first angles between the lines of the cluster point pairs and the virtual towing device and the preset constraint angle, a second score is determined according to the matching degrees between lengths of the virtual towing device and the length of the towing device, a third score is determined according to the matching degrees between lengths of the lines and the width in the size, and a fourth score is determined according to the matching degrees between the first angles between the lines of the cluster point pairs and the first angles calculated in the last frame; the first score, the second score, the third score and the fourth score are weighted and summed to obtain a first comprehensive score; the second candidate cluster point pairs are screened based on the first comprehensive score to obtain screened second candidate cluster point pairs.

7. The method according to claim 1 or 2, characterized in that, Before the prior data of the trailer and the radar point cloud collected by the robot are obtained, the method further comprises: an input page is displayed on a screen; the prior data of the trailer input in the input page are obtained.

8. The method of claim 7, wherein, Before the first angle of the trailer is determined according to the first candidate cluster point pairs, the method further comprises: a fifth score is determined according to the matching degrees between the third angles and the preset constraint angle, a sixth score is determined according to the matching degrees between lengths of the virtual towing device and the length of the towing device, a seventh score is determined according to the matching degrees between lengths of the lines and the width in the size, and an eighth score is determined according to the matching degrees between the third angles and the third angles calculated in the last frame; the fifth score, the sixth score, the seventh score and the eighth score are weighted and summed to obtain a second comprehensive score; the first candidate cluster point pairs are screened based on the second comprehensive score to obtain screened first candidate cluster point pairs.

9. The method of claim 7, wherein, The method further comprises: an outer rectangle is calculated for each sub-cluster point cloud in the first cluster point cloud; a sub-cluster point cloud with a long side of the outer rectangle less than a preset threshold is determined as the point type cluster point cloud; a sub-cluster point cloud with a long side of the outer rectangle greater than or equal to the preset threshold is determined as the line type cluster point cloud.

10. An apparatus for a robot to identify a trailer attitude, characterized by, The device comprises: An acquisition module is configured to acquire prior data of a trailer and radar point clouds collected by a robot, the prior data including a size of the trailer, a preset constraint angle between the trailer and a towing device, and a length of the towing device; A filtering module is configured to filter the radar point clouds based on a detection area of the trailer to obtain first target radar point clouds; A clustering module is configured to cluster the first target radar point clouds to obtain first clustered point clouds, the first clustered point clouds including point-type clustered point clouds and line-type clustered point clouds; A searching module is configured to search for clustered point pairs matching the prior data from the first clustered point clouds to obtain first candidate clustered point pairs; A determination module is configured to determine a first angle of the trailer according to the first candidate clustered point pairs; An identification module is configured to identify a pose of the trailer according to the prior data and the first angle; The searching for the clustered point pairs matching the prior data from the first clustered point clouds to obtain the first candidate clustered point pairs includes: combining each clustered point in the point-type clustered point clouds with each edge clustered point in the line-type clustered point clouds in pairs to obtain a plurality of clustered point pairs; determining a third angle between a connecting line of each clustered point pair in the plurality of clustered point pairs and a virtual towing device, the virtual towing device being a virtual connecting line between a midpoint of the connecting line and a towing point of the robot; and selecting a first candidate clustered point pair from the first clustered point clouds according to a matching degree between the third angle and the preset constraint angle, a matching degree between a length of the virtual towing device and a length of the towing device, a matching degree between a length of the connecting line and a width in the size, and a matching degree between the third angle and a third angle obtained in a previous frame. 11.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-10. The processor executes the computer program to implement the steps of the method in any one of claims 1 to 9.

12. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 9.

13. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 9. The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 9.

Images