Dynamic obstacle detection method, robot pose positioning method and related equipment

Abstract

The invention provides a dynamic obstacle detection method, a robot pose positioning method and related equipment, and relates to the technical field of robot automatic navigation. According to the invention, after the 2D laser radar detection data collected by the target robot in the current task scene is obtained, the obstacle distribution probability graph model is updated based on the minimum two-dimensional distances between all radar detection endpoints of the 2D laser radar detection data and the marked static obstacles in the current task scene; and then, calling the updated obstacle distribution probability graph model, and performing dynamic obstacle inference on all radar detection endpoints to determine target detection endpoints associated with dynamic obstacles in the current task scene, thereby accurately identifying 2D radar data related to the dynamic obstacles and static obstacles in the current task scene. Therefore, a high-accuracy robot pose positioning function can be realized by directly utilizing a conventional 2D laser radar positioning algorithm in a dynamic task scene.

Description

Technical Field

[0001] This application relates to the technical field of robot automatic navigation. Specifically, it relates to a method for detecting dynamic obstacles, a method for positioning the pose of a robot, and related devices. Background Art

[0002] With the continuous development of science and technology, robot technology has received extensive attention from all walks of life due to its great research value and application value. Among them, mobile robots (for example, AGV (Automated Guided Vehicle) carts commonly used in the industrial manufacturing field and floor-sweeping robots commonly used in the home service field) are an important research direction of today's robot technology. In the actual use process of mobile robots, it is often necessary to use equipped sensing devices (such as cameras, 2D lidars, ultra-wideband sensors, etc.) to instantaneously perceive the surrounding environment of the robot, so as to quickly locate the true pose of the mobile robot in the task scenario, facilitating the planning of an accurate and reliable obstacle avoidance path for the mobile robot.

[0003] However, it should be noted that in order to save the manufacturing cost of robots, major manufacturers often only equip mobile robots with 2D lidars. The effective detection data of 2D lidars usually only includes distance information. Conventional 2D lidar positioning algorithms based on 2D lidar detection data are often applicable to static task scenarios with only static obstacles and cannot be directly adapted to dynamic task scenarios with both dynamic and static obstacles. That is to say, when facing a dynamic task scenario, the robot pose data determined directly using conventional 2D lidar positioning algorithms actually has obvious data accuracy defects. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method for detecting dynamic obstacles, a method for positioning the pose of a robot, a computer device, and a readable storage medium. By introducing the probability graph model technology to classify 2D radar data, it can accurately identify the 2D radar data related to dynamic and static obstacles in the current task scenario, so as to subsequently directly use the conventional 2D lidar positioning algorithm based on the 2D radar data of static obstacles to achieve a high-accuracy robot pose positioning function, thereby effectively improving the pose positioning accuracy of mobile robots in dynamic task scenarios.

[0005] To achieve the above purpose, the technical solutions adopted in the embodiments of this application are as follows:

[0006] In the first aspect, this application provides a method for detecting dynamic obstacles, and the method includes:

[0007] Obtain the 2D lidar detection data of the target robot in the current task scenario, and determine the minimum two-dimensional distance between each radar detection endpoint of the 2D lidar detection data and the labeled static obstacles in the current task scenario;

[0008] Update the obstacle distribution probability map model based on the minimum two-dimensional distance corresponding to each of the radar detection endpoints;

[0009] Call the updated obstacle distribution probability map model to perform dynamic obstacle inference on each of the radar detection endpoints, and obtain the target detection endpoints associated with dynamic obstacles in the current task scenario.

[0010] In an optional implementation manner, the 2D lidar detection data includes multiple frames of lidar data, and each frame of lidar data corresponds to a radar detection endpoint separately. Then, the step of determining the minimum two-dimensional distance between each radar detection endpoint of the 2D lidar detection data and the labeled static obstacles in the current task scenario includes:

[0011] For each frame of lidar data, determine the actual two-dimensional coordinates of the radar detection endpoint corresponding to this frame of lidar data in the current task scenario with respect to the world coordinate system;

[0012] According to the actual two-dimensional coordinates of the radar detection endpoint and the labeled two-dimensional coordinates of all the labeled static obstacles in the two-dimensional scene map of the current task scenario, calculate the minimum two-dimensional distance corresponding to this radar detection endpoint.

[0013] In an optional implementation manner, for each frame of lidar data, the step of determining the actual two-dimensional coordinates of the radar detection endpoint corresponding to this frame of lidar data in the current task scenario with respect to the world coordinate system includes:

[0014] Obtain the original pose data of the target robot in the world coordinate system when collecting the 2D lidar detection data, where the original pose data is obtained by a 2D lidar positioning algorithm;

[0015] According to the relative pose relationship between the target robot and the 2D lidar, the radar detection direction and the effective detection distance represented by this frame of lidar data, perform geometric transformation on the original pose data to obtain the actual two-dimensional coordinates of the radar detection endpoint corresponding to this frame of lidar data.

[0016] In an optional implementation manner, the step of calculating the minimum two-dimensional distance corresponding to this radar detection endpoint according to the actual two-dimensional coordinates of the radar detection endpoint and the labeled two-dimensional coordinates of all the labeled static obstacles in the two-dimensional scene map of the current task scenario includes:

[0017] For each labeled static obstacle, calculate the Euclidean distance between the labeled two-dimensional coordinates of the labeled static obstacle and the actual two-dimensional coordinates of the radar detection endpoint;

[0018] Arrange the Euclidean distances corresponding to all the labeled static obstacles in descending order, and select the Euclidean distance with the smallest value as the minimum two-dimensional distance corresponding to the radar detection endpoint.

[0019] In an alternative embodiment, the 2D lidar detection data includes multiple frames of radar data, and each frame of radar data corresponds to a radar detection endpoint separately. Then, the step of updating the obstacle distribution probability map model based on the minimum two-dimensional distances corresponding to all the radar detection endpoints includes:

[0020] For each radar detection endpoint, use the radar detection endpoint as a graph node of the obstacle distribution probability map model, use the minimum two-dimensional distance corresponding to the radar detection endpoint as the observation variable of the corresponding graph node, and set the hidden variable of the corresponding graph node to that the radar detection endpoint belongs to a dynamic obstacle;

[0021] Initialize the likelihood distribution of the hidden variables of all the graph nodes in the obstacle distribution probability map model, and calculate the posterior probability values of the hidden variables of all the graph nodes by using the belief propagation algorithm based on the observation variables of all the graph nodes.

[0022] In an alternative embodiment, the step of calling the updated obstacle distribution probability map model to perform dynamic obstacle inference on all the radar detection endpoints respectively to obtain the target detection endpoints associated with the dynamic obstacles in the current task scenario includes:

[0023] According to the posterior probability values of the hidden variables of all the graph nodes in the obstacle distribution probability map model, perform random sampling on each graph node in the obstacle distribution probability map model to obtain the corresponding node sampling results;

[0024] For each sampled graph node in the node sampling results, compare the posterior probability value of the hidden variable of the sampled graph node with a preset probability threshold, and determine whether the radar detection endpoint corresponding to the sampled graph node belongs to the target detection endpoint according to the probability comparison result.

[0025] In an alternative embodiment, the step of determining whether the sampled graph node belongs to the target detection endpoint according to the probability comparison result includes:

[0026] In the case where the probability comparison result is that the posterior probability value of the hidden variable of the sampled graph node is greater than or equal to the preset probability threshold, determine that the radar detection endpoint corresponding to the sampled graph node belongs to the target detection endpoint;

[0027] When the probability comparison result is that the posterior probability value of the hidden variable of the sampled graph node is less than the preset probability threshold, it is determined that the radar detection endpoint corresponding to the sampled graph node does not belong to the target detection endpoint.

[0028] In a second aspect, the present application provides a method for robot pose positioning, the method including:

[0029] For the 2D lidar detection data collected by a target robot in a current task scenario, determine the target detection endpoints associated with dynamic obstacles in the 2D lidar detection data according to the dynamic obstacle detection method described in any one of the foregoing embodiments;

[0030] Perform data filtering on the lidar data corresponding to the target detection endpoints in the 2D lidar detection data, or perform a reduction processing on the influence weight of the lidar data corresponding to the target detection endpoints, to obtain processed target lidar detection data;

[0031] Based on the target lidar detection data, use a 2D lidar positioning algorithm to perform robot positioning, to obtain the actual pose data of the target robot in the world coordinate system in the current task scenario.

[0032] In a third aspect, the present application provides a computer device, including a processor and a memory, the memory storing a computer program executable by the processor, the processor being able to execute the computer program to implement the dynamic obstacle detection method described in any one of the foregoing embodiments, or the robot pose positioning method described in the foregoing embodiments.

[0033] In a fourth aspect, the present application provides a readable storage medium, on which a computer program is stored, and when the computer program is executed by a computer device, it implements the dynamic obstacle detection method described in any one of the foregoing embodiments, or the robot pose positioning method described in the foregoing embodiments.

[0034] In this case, the beneficial effects of the embodiments of the present application may include the following content:

[0035] After obtaining the 2D lidar detection data collected by the target robot in the current task scenario, this application will determine the minimum two-dimensional distance between each radar detection endpoint of the 2D lidar detection data and the static obstacles already marked in the current task scenario. Then, based on the minimum two-dimensional distance of each radar detection endpoint, the obstacle distribution probability map model will be updated, and the updated obstacle distribution probability map model will be called to infer dynamic obstacles for each radar detection endpoint respectively to determine the target detection endpoints associated with dynamic obstacles in the current task scenario, so as to accurately identify the 2D radar data related to dynamic and static obstacles in the current task scenario from the 2D lidar detection data, so that the high-accuracy robot pose positioning function can be directly realized based on the 2D radar data of static obstacles by using the conventional 2D lidar positioning algorithm, thereby improving the pose positioning accuracy of the mobile robot in the dynamic task scenario.

[0036] To make the above objects, features, and advantages of this application more obvious and understandable, the following specifically gives preferred embodiments and detailed descriptions in conjunction with the accompanying drawings as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] To more clearly illustrate the technical solutions of the embodiments of this application, the following will briefly introduce the drawings required to be used in the embodiments. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative efforts.

[0038] Figure 1 Schematic diagram of the composition of the computer device provided in the embodiment of this application;

[0039] Figure 2 Schematic diagram of the flow of the dynamic obstacle detection method provided in the embodiment of this application;

[0040] Figure 3 For Figure 2 Schematic diagram of the sub-steps included in step S210 in

[0041] Figure 4 For Figure 2 Schematic diagram of the sub-steps included in step S220 in

[0042] Figure 5 Schematic diagram of the flow of the robot pose positioning method provided in the embodiment of this application.

[0043] Reference numerals: 10 - computer device; 11 - memory; 12 - processor; 13 - communication unit. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are some, but not all, of the embodiments of this application. The components of the embodiments of this application usually described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations.

[0045] Therefore, the detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of this application without creative efforts shall fall within the scope of protection of this application.

[0046] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0047] In the description of this application, it should be understood that the orientation or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the orientation or positional relationships shown in the accompanying drawings, or the orientation or positional relationships when the product of this application is in use and commonly placed, or the orientation or positional relationships commonly understood by those skilled in the art. It is only for the convenience of describing this application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to this application.

[0048] In the description of this application, it should also be noted that unless otherwise clearly specified and defined, the terms "set", "installed", "connected", and "coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0049] In addition, in the description of the present application, it can be understood that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0050] The following will describe in detail some embodiments of the present application with reference to the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments may be combined with each other.

[0051] Please refer to Figure 1 , Figure 1 which is a schematic diagram of the composition of the computer device 10 provided by the embodiment of the present application. In the embodiment of the present application, the computer device 10 can be communicatively connected to a mobile robot that needs to achieve a pose positioning effect, and is used to obtain the 2D lidar detection data collected by the 2D lidar equipped by the mobile robot in any task scenario, and determine the specific distribution status of dynamic obstacles and static obstacles around the mobile robot in the task scenario based on the 2D lidar detection data, so as to directly utilize the conventional 2D lidar positioning algorithm based on the 2D radar data related to the static obstacles in the 2D lidar detection data to implement a high-accuracy robot pose positioning function, that is, accurately locate the actual pose data of the mobile robot in the world coordinate system in the corresponding task scenario. Among them, the mobile robot can be, but is not limited to: an AGV cart equipped with at least a 2D lidar, a floor cleaning robot equipped with at least a 2D lidar, etc.; the computer device 10 can be a computing device independent of the mobile robot, or a hardware module device integrated with the mobile robot, where the computing device can be, but is not limited to: a personal computer, a cloud server, a laptop computer, a tablet computer, etc.

[0052] In an embodiment of the present application, the computer device 10 may include a memory 11, a processor 12, and a communication unit 13. Among them, each element of the memory 11, the processor 12, and the communication unit 13 is electrically connected directly or indirectly to each other to achieve data transmission or interaction. For example, these elements of the memory 11, the processor 12, and the communication unit 13 may be electrically connected to each other through one or more communication buses or signal lines.

[0053] In an embodiment of the present application, the memory 11 may be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. Among them, the memory 11 is used to store a computer program, and after receiving an execution instruction, the processor 12 can execute the computer program accordingly. In addition, the memory 11 is also used to store two-dimensional scene maps for different task scenarios, where each two-dimensional scene map is marked with the marked two-dimensional coordinates of all static obstacles in the corresponding task scenario relative to the world coordinate system.

[0054] In an embodiment of the present application, the processor 12 may be an integrated circuit chip with signal processing capabilities. The processor 12 may be a general-purpose processor, including at least one of a central processing unit (CPU), a graphics processing unit (GPU), a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc., and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application.

[0055] In an embodiment of the present application, the communication unit 13 is configured to establish a communication connection between the computer device 10 and other electronic devices through a network, and transmit and receive data through the network, where the network includes a wired communication network and a wireless communication network. For example, the computer device 10 can obtain 2D lidar detection data collected by a mobile robot through the communication unit 13.

[0056] Optionally, in an embodiment of the present application, the computer device 10 may pre-store a specific computer program related to the dynamic obstacle detection function in the memory 11, and by driving the processor 12 to execute the specific computer program correspondingly, based on the 2D lidar detection data collected by the mobile robot, introduce a probability map model technology to classify the 2D lidar data, so as to accurately identify the 2D lidar data related to the dynamic obstacles and static obstacles around the robot in the corresponding task scenario, so that subsequently, based on the 2D lidar data of the static obstacles, the conventional 2D lidar positioning algorithm can be directly used to achieve a high-accuracy robot pose positioning function, thereby improving the pose positioning accuracy of the mobile robot in the dynamic task scenario.

[0057] Optionally, in an embodiment of the present application, the computer device 10 may pre-store a specific computer program related to the robot pose positioning function in the memory 11, and by driving the processor 12 to execute the specific computer program correspondingly, based on the identification of the 2D lidar data related to the dynamic obstacles and static obstacles in the corresponding task scenario, weaken the influence of the 2D lidar data of the dynamic obstacles in the robot pose positioning process, and be able to directly use the conventional 2D lidar positioning algorithm based on the 2D lidar data of the static obstacles to achieve a high-accuracy robot pose positioning function, thereby effectively improving the pose positioning accuracy of the mobile robot in the dynamic task scenario.

[0058] It can be understood that Figure 1 the block diagram shown is only a schematic diagram of the composition of the computer device 10, and the computer device 10 may further include more or fewer components than those shown Figure 1 in it, or have a different configuration from that shown Figure 1 in it. Figure 1 Each component shown in it can be implemented by hardware, software, or a combination thereof.

[0059] In this application, to ensure that the computer device 10 can accurately identify the 2D radar data related to dynamic and static obstacles in the task scenario where the robot is located by introducing the probability graph model technology for 2D radar data classification, and to facilitate the subsequent realization of the high-accuracy robot pose positioning function using the conventional 2D lidar positioning algorithm, the embodiments of this application achieve the foregoing objective by providing a dynamic obstacle detection method. The dynamic obstacle detection method provided by this application is described in detail below.

[0060] Please refer to Figure 2 , Figure 2 which is a schematic flowchart of the dynamic obstacle detection method provided by the embodiments of this application. In the embodiments of this application, the dynamic obstacle detection method may include steps S210 to S230.

[0061] Step S210: Obtain the 2D lidar detection data of the target robot in the current task scenario, and determine the minimum two-dimensional distance between each radar detection endpoint of the 2D lidar detection data and the static obstacles already marked in the current task scenario.

[0062] In this embodiment, the target robot is a movable robot communicatively connected to the computer device 10, and the 2D lidar detection data is used to describe the environmental conditions around the target robot; during the process of the computer device 10 obtaining the 2D lidar detection data collected by the target robot in the current task scenario, the original pose data of the target robot in the world coordinate system when collecting the 2D lidar detection data is synchronously obtained, so as to determine the actual two-dimensional coordinates of each radar detection endpoint of the 2D lidar detection data relative to the world coordinate system through geometric relationship transformation based on the original pose data and the 2D lidar detection data (which includes multiple frames of radar data, each frame of radar data corresponding to a single radar detection direction and recording the effective detection distance between the 2D lidar and the detected object in the corresponding radar detection direction), and then call the two-dimensional scene map adapted to the current task scenario, and calculate the minimum two-dimensional distance of each radar detection endpoint in the current task scenario (i.e., the straight-line distance from the corresponding radar detection endpoint to the nearest static obstacle in the current task scenario) according to the actual two-dimensional coordinates of each radar detection endpoint and the marked two-dimensional coordinates of each static obstacle already marked in the two-dimensional scene map.

[0063] During this process, each radar detection endpoint in the 2D lidar detection data corresponds to a frame of radar data separately, and each radar detection endpoint is the actual detection position of the 2D lidar on the detected object; the original pose data can be determined by a conventional 2D lidar positioning algorithm before the 2D lidar detection data is collected, or can be predicted by using other auxiliary algorithms or auxiliary devices. The present application does not limit the specific acquisition method of the original pose data.

[0064] Optionally, please refer to Figure 3 , Figure 3 is Figure 2 a schematic flowchart of the sub-steps included in step S210 in. In the embodiment of the present application, the 2D lidar detection data includes multiple frames of radar data, and each frame of radar data corresponds to a radar detection endpoint separately. The step of "determining the minimum two-dimensional distance between all radar detection endpoints of the 2D lidar detection data and the static obstacles marked in the current task scenario" in step S210 may include sub-steps S211 to S212 to accurately measure the straight-line distance from each radar detection endpoint of the 2D lidar detection data to the nearest static obstacle.

[0065] Sub-step S211: For each frame of radar data, determine the actual two-dimensional coordinates of the radar detection endpoint corresponding to this frame of radar data in the current task scenario relative to the world coordinate system.

[0066] In this embodiment, for any frame of radar data included in the 2D lidar detection data, the computer device 10 can perform a geometric transformation on the original pose data according to the relative pose relationship between the target robot and the 2D lidar, the radar detection direction and the effective detection distance represented by this frame of radar data, to obtain the actual two-dimensional coordinates of the radar detection endpoint corresponding to this frame of radar data.

[0067] Optionally, in an implementation manner of this embodiment, for each frame of radar data, the step of determining the actual two-dimensional coordinates of the radar detection endpoint corresponding to this frame of radar data in the current task scenario relative to the world coordinate system may include: obtaining the original pose data of the target robot in the world coordinate system when collecting the 2D lidar detection data, where the original pose data is obtained by positioning through a 2D lidar positioning algorithm; performing a geometric transformation on the original pose data according to the relative pose relationship between the target robot and the 2D lidar, the radar detection direction and the effective detection distance represented by this frame of radar data, to obtain the actual two-dimensional coordinates of the radar detection endpoint corresponding to this frame of radar data.

[0068] Sub-step S212: Calculate the minimum two-dimensional distance corresponding to the radar detection endpoint based on the actual two-dimensional coordinates of the radar detection endpoint and the labeled two-dimensional coordinates of all the labeled static obstacles in the two-dimensional scene map of the current task scenario.

[0069] In this embodiment, after determining the actual two-dimensional coordinates of a certain radar detection endpoint, the computer device 10 can calculate the Euclidean distance between the labeled two-dimensional coordinates of the labeled static obstacle and the actual two-dimensional coordinates of the radar detection endpoint for each labeled static obstacle in the current task scenario, and then sort the Euclidean distances corresponding to all the labeled static obstacles in the current task scenario in descending order, and select the Euclidean distance with the smallest value as the minimum two-dimensional distance corresponding to the radar detection endpoint, so as to measure the straight-line distance from the radar detection endpoint to the nearest static obstacle in the current task scenario.

[0070] Thus, the present application can accurately measure the straight-line distances from each radar detection endpoint of the 2D lidar detection data to the nearest static obstacle by executing the above sub-steps S211 to S212.

[0071] Step S220: Update the obstacle distribution probability map model based on the minimum two-dimensional distances corresponding to all the radar detection endpoints.

[0072] In this embodiment, the obstacle distribution probability map model is used to predict whether each frame of radar data in the 2D lidar detection data belongs to a dynamic obstacle, that is, to determine the specific distribution of dynamic obstacles and static obstacles around the target robot in the current task scenario. Among them, each graph node in the obstacle distribution probability map model corresponds to a radar detection endpoint alone, the observation variable of each graph node in the obstacle distribution probability map model is represented by "the minimum two-dimensional distance corresponding to the radar detection endpoint", and the hidden variable of each graph node in the obstacle distribution probability map model is represented by "the radar detection endpoint belongs to a dynamic obstacle".

[0073] At this time, reference can be made to Figure 4 , Figure 4 Yes Figure 2 is a schematic flowchart of the sub-steps included in step S220 in

[0074] Sub-step S221: For each radar detection endpoint, use this radar detection endpoint as a graph node of the obstacle distribution probability map model, and use the minimum two-dimensional distance corresponding to this radar detection endpoint as the observation variable of the corresponding graph node. At the same time, set the hidden variable of the corresponding graph node to that this radar detection endpoint belongs to a dynamic obstacle.

[0075] Sub-step S222: Initialize the likelihood distribution of the hidden variables of all graph nodes in the obstacle distribution probability map model, and calculate the posterior probability values of the hidden variables of all graph nodes based on the observation variables of all graph nodes using the belief propagation algorithm.

[0076] Step S230: Invoke the updated obstacle distribution probability map model to perform dynamic obstacle inference on all radar detection endpoints respectively, and obtain the target detection endpoints associated with dynamic obstacles in the current task scenario.

[0077] In this embodiment, after the computer device 10 completes the update operation of the obstacle distribution probability map model, it can perform random sampling on each graph node in the obstacle distribution probability map model according to the posterior probability values of the hidden variables of all graph nodes in the obstacle distribution probability map model to obtain the corresponding node sampling results. Then, for each sampled graph node in the node sampling results, compare the posterior probability value of the hidden variable of this sampled graph node with a preset probability threshold, and judge whether the radar detection endpoint corresponding to this sampled graph node belongs to the target detection endpoint according to the probability comparison result, so as to complete the dynamic obstacle inference process of all radar detection endpoints. Among them, for any sampled graph node, when the probability comparison result of this sampled graph node is "the posterior probability value of the hidden variable of this sampled graph node is greater than or equal to the preset probability threshold", it can be determined that the radar detection endpoint corresponding to this sampled graph node belongs to the target detection endpoint, and when the probability comparison result of this sampled graph node is "the posterior probability value of the hidden variable of this sampled graph node is less than the preset probability threshold", it can be determined that the radar detection endpoint corresponding to this sampled graph node does not belong to the target detection endpoint.

[0078] In this embodiment, when a radar detection endpoint is inferred as a target detection endpoint, it indicates that the 2D radar data corresponding to this radar detection endpoint in the 2D lidar detection data involves a dynamic obstacle. When the computer device 10 determines all the target detection endpoints of the 2D lidar detection data by invoking the updated obstacle distribution probability map model, it indicates that the 2D radar data involved in the dynamic and static obstacles around the robot in the corresponding task scenario has been accurately identified. At this time, the computer device 10 can directly use the conventional 2D lidar positioning algorithm for robot positioning based on the radar data associated with the static obstacle in the 2D lidar detection data, and obtain the actual pose data of the target robot in the world coordinate system in the current task scenario, thereby realizing a high-accuracy robot pose positioning function in the dynamic task scenario.

[0079] Therefore, this application can perform the above steps S210 to S230, classify the 2D radar data by introducing the probability map model technology on the basis of the 2D lidar detection data collected by the mobile robot, and accurately identify the 2D radar data involved in the dynamic and static obstacles around the robot in the corresponding task scenario, so as to directly use the conventional 2D lidar positioning algorithm to realize a high-accuracy robot pose positioning function according to the 2D radar data of the static obstacle, and improve the pose positioning accuracy of the mobile robot in the dynamic task scenario.

[0080] In this application, to ensure that the computer device 10 can, on the basis of identifying the 2D radar data involved in the dynamic and static obstacles in the corresponding task scenario, weaken the influence of the 2D radar data of the dynamic obstacle in the robot pose positioning process, and use the conventional 2D lidar positioning algorithm to realize a high-accuracy robot pose positioning function in the dynamic task scenario, the embodiment of this application achieves the foregoing purpose by providing a robot pose positioning method. The robot pose positioning method provided by this application will be described in detail below.

[0081] Please refer to Figure 5 , Figure 5 which is a schematic flowchart of the robot pose positioning method provided by the embodiment of this application. In the embodiment of this application, the robot pose positioning method may include steps S310 to S330.

[0082] Step S310: For the 2D lidar detection data collected by the target robot in the current task scenario, determine the target detection endpoints associated with the dynamic obstacle in the 2D lidar detection data according to the dynamic obstacle detection method.

[0083] In this embodiment, the dynamic obstacle detection method is the dynamic obstacle detection method described above, so as to accurately identify the 2D radar data related to the dynamic obstacles and static obstacles around the target robot in the current task scenario.

[0084] Step S320: Perform data filtering on the radar data corresponding to the target detection end point in the 2D lidar detection data, or perform a reduction in the influence weight on the radar data corresponding to the target detection end point, to obtain the processed target radar detection data.

[0085] In this embodiment, when performing a data filtering operation on the 2D lidar detection data, the corresponding target radar detection data is composed of the radar data in the 2D lidar detection data associated with the labeled static obstacles.

[0086] When performing a reduction in the influence weight on the radar data corresponding to the target detection end point, although the target radar detection data is still represented by the 2D lidar detection data, the radar data corresponding to the target detection end point in the target radar detection data will maintain a low influence weight state during the robot pose positioning process, while the radar data associated with the labeled static obstacles will still maintain the original high influence weight state during the robot pose positioning process, so as to weaken the influence of the 2D radar data of the dynamic obstacles in the robot pose positioning process using the conventional 2D lidar positioning algorithm.

[0087] Step S330: Based on the target radar detection data, use the 2D lidar positioning algorithm to perform robot positioning, to obtain the actual pose data of the target robot in the world coordinate system in the current task scenario.

[0088] Thus, by executing the above steps S310 to S330, the present application can, on the basis of identifying the 2D radar data related to the dynamic obstacles and static obstacles in the corresponding task scenario, weaken the influence of the 2D radar data of the dynamic obstacles in the robot pose positioning process, and can directly use the conventional 2D lidar positioning algorithm to achieve a high-accuracy robot pose positioning function based on the 2D radar data of the static obstacles, thereby effectively improving the pose positioning accuracy of the mobile robot in the dynamic task scenario.

[0089] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of the devices, methods, and computer program products according to the embodiments of this application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and a module, a program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0090] In addition, the functional modules in each embodiment of this application can be integrated together to form an independent part, or each module can exist alone, or two or more modules can be integrated to form an independent part. If the various functions provided in this application are implemented in the form of software functional modules and sold or used as an independent product, they can be stored in a storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (such as a laptop computer, a sweeping robot, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned readable storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0091] As described above, these are only the various implementation manners of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A dynamic obstacle detection method, characterized in that: The method comprises: Obtain 2D laser radar detection data of the target robot in the current task scene, and determine the minimum two-dimensional distance between all radar detection endpoints of the 2D laser radar detection data and the marked static obstacles in the current task scene; Based on the minimum two-dimensional distances corresponding to all the radar detection endpoints, the obstacle distribution probability graph model is updated; The updated obstacle distribution probability graph model is called to perform dynamic obstacle inference on all the radar detection endpoints to obtain the target detection endpoints associated with the dynamic obstacles in the current mission scenario.

2. The method according to claim 1, characterized in that The 2D laser radar detection data includes multiple frames of radar data, each frame of radar data corresponds to a radar detection endpoint, and the step of determining the minimum two-dimensional distance between all radar detection endpoints of the 2D laser radar detection data and the marked static obstacles in the current task scene includes: For each frame of radar data, determine the actual two-dimensional coordinates of the radar detection endpoint corresponding to the frame of radar data relative to the world coordinate system in the current mission scene; According to the actual two-dimensional coordinates of the radar detection endpoint and the marked two-dimensional coordinates of all marked static obstacles in the two-dimensional scene map of the current mission scene, the minimum two-dimensional distance corresponding to the radar detection endpoint is calculated.

3. The method according to claim 2, characterized in that For each frame of radar data, the step of determining the actual two-dimensional coordinates of the radar detection endpoint corresponding to the frame of radar data relative to the world coordinate system in the current mission scene includes: Acquire original position data of the target robot in the world coordinate system when the target robot collects the 2D laser radar detection data, wherein the original position data is obtained by positioning using a 2D laser radar positioning algorithm; According to the relative posture relationship between the target robot and the 2D laser radar, the radar detection direction and effective detection distance represented by the frame radar data, the original posture data is geometrically transformed to obtain the actual two-dimensional coordinates of the radar detection endpoint corresponding to the frame radar data.

4. The method according to claim 2, characterized in that: The step of calculating the minimum two-dimensional distance corresponding to the radar detection endpoint according to the actual two-dimensional coordinates of the radar detection endpoint and the marked two-dimensional coordinates of all marked static obstacles in the two-dimensional scene map of the current task scene includes: For each marked static obstacle, calculating the Euclidean distance between the marked two-dimensional coordinates of the marked static obstacle and the actual two-dimensional coordinates of the radar detection endpoint; The Euclidean distances corresponding to all marked static obstacles are arranged in descending order, and the Euclidean distance with the smallest value is selected as the minimum two-dimensional distance corresponding to the radar detection endpoint.

5. The method according to any one of claims 1 to 4, characterized in that: The 2D laser radar detection data includes multiple frames of radar data, each frame of radar data corresponds to a radar detection endpoint, and the step of updating the obstacle distribution probability graph model based on the minimum two-dimensional distances corresponding to all the radar detection endpoints includes: For each radar detection endpoint, the radar detection endpoint is used as a graph node of the obstacle distribution probability graph model, and the minimum two-dimensional distance corresponding to the radar detection endpoint is used as the observed variable of the corresponding graph node, and the hidden variable of the corresponding graph node is set to indicate that the radar detection endpoint belongs to a dynamic obstacle; The likelihood distribution of the hidden variables of all the graph nodes in the obstacle distribution probability graph model is initialized, and based on the observed variables of all the graph nodes, the posterior probability values ​​of the hidden variables of all the graph nodes are calculated using the belief propagation algorithm.

6. The method according to claim 5, characterized in that The step of calling the updated obstacle distribution probability graph model, performing dynamic obstacle inference on all radar detection endpoints respectively, and obtaining target detection endpoints associated with dynamic obstacles in the current mission scenario includes: According to the posterior probability values ​​of the hidden variables of all the graph nodes in the obstacle distribution probability graph model, randomly sampling each graph node in the obstacle distribution probability graph model to obtain corresponding node sampling results; For each sampled graph node in the node sampling result, the posterior probability value of the hidden variable of the sampled graph node is compared with a preset probability threshold, and according to the probability comparison result, it is determined whether the radar detection endpoint corresponding to the sampled graph node belongs to the target detection endpoint.

7. The method according to claim 6, characterized in that The step of judging whether the sampled graph node belongs to the target detection endpoint according to the probability comparison result includes: When the probability comparison result is that the posterior probability value of the hidden variable of the sampled graph node is greater than or equal to the preset probability threshold, determining that the radar detection endpoint corresponding to the sampled graph node belongs to the target detection endpoint; When the probability comparison result is that the posterior probability value of the hidden variable of the sampled graph node is less than the preset probability threshold, it is determined that the radar detection endpoint corresponding to the sampled graph node does not belong to the target detection endpoint.

8. A robot posture positioning method, characterized in that: The method comprises: For the 2D laser radar detection data collected by the target robot in the current task scenario, determine the target detection endpoint associated with the dynamic obstacle in the 2D laser radar detection data according to the dynamic obstacle detection method described in any one of claims 1 to 7; Performing data filtering processing on the radar data corresponding to the target detection endpoint in the 2D laser radar detection data, or performing influence weight reduction processing on the radar data corresponding to the target detection endpoint, to obtain processed target radar detection data; Based on the target radar detection data, the robot is positioned using a 2D lidar positioning algorithm to obtain the actual position data of the target robot in the world coordinate system in the current task scenario.

9. A computer device, characterized in that: It includes a processor and a memory, the memory stores a computer program that can be executed by the processor, and the processor can execute the computer program to implement the dynamic obstacle detection method described in any one of claims 1 to 7, or the robot posture positioning method described in claim 8.

10. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a computer device, the dynamic obstacle detection method described in any one of claims 1 to 7, or the robot posture positioning method described in claim 8 is implemented.

Images