A control method, device and electronic equipment for unmanned equipment

Abstract

This specification discloses a control method, device and electronic device for unmanned equipment. The embodiments of this specification determine the expansion distance of the obstacle based on the system error of the driving system of the unmanned equipment, and at the same time, determine the expansion direction of the obstacle's outline based on the relative position between the obstacle's current initial planned path and the obstacle. Then, the outline of the obstacle is expanded according to the expansion direction and the expansion distance, and the initial planned path is adjusted based on the expanded outline. The unmanned equipment is controlled to travel according to the adjusted planned path. In this method, the outline of the obstacle is expanded in a certain direction and distance, so that the passable area used to plan the driving path of the unmanned equipment becomes smaller. In this way, the amplitude of adjusting the initial planned path can be reduced, thereby reducing the frequency of changing the driving path of the unmanned equipment.

Description

Technical Field

[0001] This specification relates to the field of unmanned driving, and in particular to a control method, device and electronic equipment for unmanned equipment. Background Art

[0002] In the field of unmanned driving, the automatic driving of unmanned equipment involves the collaboration of various systems, and ultimately the control system controls the unmanned equipment to drive safely. Various systems may include: perception system, positioning system, prediction system, planning system, control system, etc. Among them, the perception system is used to perceive the position of obstacles around the unmanned equipment, the positioning system is used to locate the position of the unmanned equipment itself, the prediction system is used to predict the future movement path of obstacles, the planning system is used to plan the movement path of the unmanned equipment itself, and the control system is used to control the unmanned equipment to move according to the movement path planned by the unmanned equipment.

[0003] Taking the planning system as an example, when planning the path of unmanned equipment, the planning system periodically integrates the data obtained from various systems such as the positioning system, perception system, and prediction system, and plans the motion path corresponding to each cycle for the unmanned equipment based on the data obtained from each system.

[0004] However, when the positioning system, perception system, prediction system and other systems obtain the corresponding data, due to the certain errors in each system itself, the movement path planned by the planning system in each cycle has a large deviation, causing the unmanned equipment to frequently change its movement trajectory during driving. Therefore, how to reduce the deviation between the movement paths planned by the planning system in each cycle is an urgent problem to be solved. Summary of the invention

[0005] The embodiments of this specification provide a control method, device and electronic device for unmanned equipment to at least partially solve the problems existing in the above-mentioned prior art.

[0006] The embodiments of this specification adopt the following technical solutions:

[0007] This specification provides a control method for unmanned equipment, including:

[0008] Obtaining an initial planned path currently corresponding to the unmanned device and a system error of a driving system used by the unmanned device;

[0009] Determining the expansion direction of the outline of the obstacle according to the relative position between the initial planned path and the obstacle within the preset range of the unmanned equipment;

[0010] Determine, according to the system error, an expansion distance corresponding to the outline of the obstacle, wherein the larger the system error is, the larger the expansion distance is;

[0011] Expanding the outline of the obstacle according to the expansion distance and the expansion direction to obtain the expanded outline of the obstacle;

[0012] The initial planned path is adjusted according to the expanded outline of the obstacle to obtain an adjusted planned path, and the unmanned equipment is controlled to travel according to the adjusted planned path.

[0013] Optionally, the driving system includes: at least one of a perception system, a positioning system, a prediction system, and a control system;

[0014] The system error of the perception system is a perception error, and the perception error refers to the error between the position of the obstacle perceived by the perception system and the actual position of the obstacle;

[0015] The system error of the positioning system is the positioning error, and the positioning error refers to the error between the positioning position of the unmanned equipment determined by the positioning system and the actual position of the unmanned equipment;

[0016] The system error of the prediction system is the prediction error, and the prediction error refers to the error between the trajectory position of the obstacle predicted by the prediction system and the actual position of the obstacle;

[0017] The system error of the control system is a control error, and the control error refers to the error between the target position that the unmanned equipment should reach under the control system and the positioning position of the unmanned equipment.

[0018] Optionally, determining the expansion direction of the outline of the obstacle according to the relative position between the initial planned path and the obstacle within a preset range of the unmanned equipment includes:

[0019] Based on the edge on the outline of the obstacle, the area around the obstacle is divided to obtain a plurality of sub-areas; wherein one sub-area corresponds to one edge on the outline;

[0020] According to the position of the initial planned path and the area positions of the multiple sub-areas, determining a sub-area that does not cover at least a portion of the initial planned path from the multiple sub-areas as a target sub-area;

[0021] The direction in which the edge corresponding to the target sub-region is away from the geometric center of the obstacle is used as the expansion direction of the outline of the obstacle.

[0022] Optionally, based on the edge on the outline of the obstacle, the area around the obstacle is divided to obtain multiple sub-areas, specifically including:

[0023] For each edge on the outline of the obstacle, determine a line between the geometric center of the obstacle and the endpoint of the edge as the line corresponding to the edge;

[0024] A plurality of regions enclosed by the intersection of the extended line of the connecting line corresponding to each edge and the road boundary of the road region where the obstacle is located are used as a plurality of sub-regions.

[0025] Optionally, determining the expansion distance corresponding to the outline of the obstacle according to the system error includes:

[0026] The system error is input into a pre-trained prediction model, so that the prediction model predicts the expansion distance corresponding to the obstacle according to the regression relationship between the system error and the sample error used in the prediction model training process.

[0027] Optionally, before inputting the system error into a pre-trained prediction model, the method further comprises:

[0028] Obtain the system error of the driving system used by the unmanned equipment in history as the sample error;

[0029] Inputting each sample error into the prediction model to be trained, so that the prediction model predicts the expansion distance to be optimized corresponding to the contour of the historical obstacle according to the regression relationship between each sample error;

[0030] The prediction model is trained with the goal of minimizing the difference between the expansion distance to be optimized and the actual expansion distance corresponding to the contour of the historical obstacle.

[0031] Optionally, determining the expansion distance corresponding to the outline of the obstacle according to the system error includes:

[0032] Determining an initial expansion distance interval corresponding to the outline of the obstacle according to the system error;

[0033] Dividing the initial expansion distance interval into equal intervals to obtain a plurality of sub-distance intervals;

[0034] By means of the simulation system corresponding to the unmanned equipment, for each sub-distance interval, the expanded distance corresponding to the sub-distance interval is evaluated according to the driving path adjusted after the obstacle expands the expanded distance corresponding to the sub-distance interval, so as to obtain an evaluation result corresponding to the sub-distance interval;

[0035] According to the evaluation result corresponding to each sub-distance interval, the expansion distance corresponding to the outline of the obstacle is selected from the initial expansion distance interval.

[0036] This specification provides an unmanned equipment control device, including:

[0037] An acquisition module is used to acquire an initial planned path currently corresponding to the unmanned equipment and a system error of a driving system used by the unmanned equipment;

[0038] A first determination module, configured to determine an expansion direction of a contour of the obstacle according to a relative position between the initial planned path and an obstacle within a preset range of the unmanned equipment;

[0039] A second determination module determines, according to the system error, an expansion distance corresponding to the outline of the obstacle, wherein the larger the system error is, the larger the expansion distance is;

[0040] An expansion module, used to expand the outline of the obstacle according to the expansion distance and the expansion direction to obtain the expanded outline of the obstacle;

[0041] The control module is used to adjust the initial planned path according to the expanded outline of the obstacle to obtain an adjusted planned path, and control the unmanned equipment to travel according to the adjusted planned path.

[0042] This specification provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the control method of the unmanned equipment described above is implemented.

[0043] The present specification provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned control method for unmanned equipment when executing the program.

[0044] At least one of the above technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects:

[0045] The embodiments of this specification determine the expansion distance of the obstacle based on the system error of the driving system of the unmanned equipment, and at the same time, determine the expansion direction of the outline of the obstacle based on the relative position between the current initial planned path of the obstacle and the obstacle. Then, the outline of the obstacle is expanded according to the expansion direction and the expansion distance, and the initial planned path is adjusted based on the expanded outline. The unmanned equipment is controlled to travel according to the adjusted path. In this method, the outline of the obstacle is expanded in a certain direction and distance, so that the passable area used to plan the driving path of the unmanned equipment becomes smaller. In this way, the amplitude of adjusting the initial planned path can be reduced, thereby reducing the frequency of changing the driving path of the unmanned equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings described herein are used to provide a further understanding of this specification and constitute a part of this specification. The illustrative embodiments and descriptions of this specification are used to explain this specification and do not constitute an improper limitation on this specification. In the drawings:

[0047] Figure 1a to Figure 1b A schematic diagram of the paths planned for the first cycle and the second cycle provided by the prior art;

[0048] Figure 2 A schematic diagram of the control flow of unmanned equipment provided in the embodiments of this specification;

[0049] Figure 3 A schematic diagram of the relative positions of the initial planned path and obstacles provided in the embodiments of this specification;

[0050] Figure 4a to Figure 4b A schematic diagram of the regional division provided in the embodiments of this specification;

[0051] Figure 5 A schematic diagram of an expanded outline of an obstacle provided in an embodiment of this specification;

[0052] Figure 6 A schematic diagram of the adjusted initial planned path provided in the embodiments of this specification;

[0053] Figure 7 A schematic diagram of the structure of an unmanned equipment control device provided in an embodiment of this specification;

[0054] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0055] In the prior art, due to the system errors of each system in the driving system of the unmanned equipment, the limited computing power of the system, the instability of the environment around the unmanned equipment, etc., each system in the driving system needs to determine the relevant information of the obstacles around the unmanned equipment and periodically plan the driving path for the unmanned equipment. Among them, the periodic planning of the driving path refers to the real-time planning of the driving path of the unmanned equipment within a certain period of time in the future.

[0056] Taking the planning system as an example, when planning the future path of the unmanned equipment, the planning system needs to integrate the data obtained by other systems in the driving system. Other systems may include: perception system, positioning system, prediction system, map system, control system, etc. The map system is used to determine the locations of obstacles and unmanned equipment in the map.

[0057] Due to the system errors of other systems in the driving system, the data obtained by other systems have errors. In this way, there will naturally be certain errors when the planning system plans the driving path of the unmanned equipment. Since the planning system needs to periodically plan the driving path of the unmanned equipment and there will be certain errors when planning the path in each cycle, the driving paths planned by the planning system in the same time period in each cycle may be greatly different. Because the driving paths planned by the planning system for the same time period in the future may be greatly different in each cycle, this will cause the unmanned equipment to frequently change its driving trajectory during actual driving, causing the unmanned equipment to swerve.

[0058] Take the paths planned by the first and second cycles of the planning system as an example. Figure 1a The driving path planned for the first cycle provided by the prior art. Figure 1b The driving path planned by the second cycle provided by the prior art, wherein a cycle is taken as an example of 20 seconds, that is, a cycle may refer to a time period from the current moment to the next 20 seconds.

[0059] exist Figure 1a to Figure 1b In the example, there are unmanned equipment, obstacle A, obstacle B, and the driving path planned by the unmanned equipment within the road boundary. Figure 1a In the figure, the driving path planned by the unmanned equipment is between obstacle A and the road boundary. This is the driving path planned by the planning system for the unmanned equipment in the first cycle. Due to the error of the driving system itself, the driving path planned by the unmanned equipment in the second cycle has a large deviation from the driving path planned in the first cycle, such as Figure 1b As shown. Figure 1b In the figure, the driving path planned by the unmanned equipment is located between obstacle A and obstacle B.

[0060] In the prior art, the system errors of each system in the driving system can easily lead to large deviations in the driving path planned by the planning system in each cycle, and the most basic condition for planning a path is to plan the path in an area where unmanned equipment can pass safely. Therefore, in order to reduce the problem of large deviations in the driving path planned in each cycle, the area where unmanned equipment can pass is compressed in this specification, that is, the outline of the obstacle is expanded, so as to reduce the possibility of a sudden change in the path planned in the current cycle based on the path planned in the previous cycle.

[0061] Therefore, the control method of the unmanned equipment provided in this specification is intended to directionally expand the outline of the obstacle, thereby compressing the drivable area of ​​the unmanned equipment.

[0062] In order to make the purpose, technical solutions and advantages of this specification more clear, the technical solutions of this specification will be clearly and completely described below in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this specification.

[0063] The technical solutions provided by the embodiments of this specification are described in detail below in conjunction with the accompanying drawings.

[0064] Figure 2 The control flow diagram of the unmanned equipment provided in the embodiment of this specification includes:

[0065] S200: Acquire the initial planned path currently corresponding to the unmanned equipment and the system error of the driving system used by the unmanned equipment.

[0066] In the embodiments of this specification, the driving system of the unmanned equipment may include various systems such as a perception system, a positioning system, a prediction system, a planning system, a control system, and a map system. In addition, each system in the driving system corresponds to its own system error. For example: the system error of the perception system is the perception error, and the perception error refers to the error between the position of the obstacle perceived by the perception system and the actual position of the obstacle. The system error of the positioning system is the positioning error, and the positioning error refers to the error between the positioning position of the unmanned equipment determined by the positioning system and the actual position of the unmanned equipment. The system error of the prediction system is the prediction error, and the prediction error refers to the error between the trajectory position of the obstacle predicted by the prediction system and the actual position of the obstacle. The system error of the control system is the control error, and the control error refers to the error between the target position that the control system controls the unmanned equipment to reach and the positioning position of the unmanned equipment.

[0067] In the embodiment of the present specification, the initial planned path currently corresponding to the unmanned device can be obtained through the planning system in the driving system of the unmanned device. The initial planned path currently corresponding to the unmanned device is at least part of the driving path planned by the planning system at the last moment.

[0068] At the same time, the system error of at least one system in the driving system can be obtained through the driving system used by the unmanned equipment.

[0069] It should be noted that the unmanned equipment in this specification can be unmanned vehicles and drones. The unmanned equipment can be used in the field of logistics and distribution, including both instant delivery fields such as takeout and delivery, and other non-instant delivery fields. Unmanned equipment can also be used in manned services. In addition, Figure 2The control method of the unmanned equipment shown is applied to the unmanned equipment, and can also be applied to the server that controls the unmanned equipment.

[0070] S202: Determine an expansion direction of a contour of the obstacle according to a relative position between the initial planned path and an obstacle within a preset range of the unmanned equipment.

[0071] In the embodiments of this specification, it is necessary to compress the drivable area of ​​the current unmanned equipment, that is, to directionally expand the outline of the obstacle. In this way, the problem of large differences between the currently planned driving path and the initial planned path can be reduced. Among them, the obstacle can be an object other than the unmanned equipment itself, which can be a manned vehicle, a pedestrian, a static object, etc.

[0072] In the perception system, obstacle data within a preset range of the unmanned equipment can be collected by sensors, wherein the data type of the obstacle data may include image data and point cloud data. Then, obstacles and the outline of obstacles are determined based on the image data and / or point cloud data collected by the sensor. The outline of the obstacle may be a polygon formed by projecting the obstacle onto the ground.

[0073] Specifically, the point cloud data of obstacles within the preset range of the unmanned equipment can be collected by laser radar, and the collected point cloud data can be clustered according to the position of each point cloud point to obtain multiple clusters. Among them, each cluster can represent an obstacle. Then, for each cluster, the peripheral point cloud points corresponding to the cluster are determined, and the peripheral point cloud points corresponding to the cluster are projected to the ground plane to determine the contour corresponding to the cluster. That is, the contour of the obstacle is determined.

[0074] In addition, the image data of obstacles within the preset range of the unmanned equipment can also be collected by the visual sensor. The image data of the obstacle can be the overhead image data of the obstacle relative to the ground. Then, the collected image data is subjected to image recognition to identify the obstacles in the image. Then, the outline of each obstacle is extracted according to the pixel points of each obstacle in the image.

[0075] The outline of the obstacle may be the outer contour of the obstacle projected onto the ground, or may be the minimum polygon containing the obstacle extracted by using a polygon extraction frame.

[0076] At the same time, the current position of each obstacle can be determined based on the positioning position of the unmanned equipment and the image data and / or point cloud data corresponding to each obstacle. Then, the expansion direction of the obstacle outline is determined based on the relative position between the currently corresponding initial planning path and the current position of the obstacle. The relative position between the initial planning path and the current position of the obstacle can refer to the direction of the obstacle where the initial planning path is located, for example: taking the driving direction of the unmanned equipment as the front of the obstacle, the initial planning path is at least partially located on the left side of the obstacle or the initial planning path is at least partially located on the right side of the obstacle.

[0077] Specifically, the current position of the unmanned device in the map and the position of each obstacle in the map can be determined based on the current positioning position of the unmanned device and the position of each obstacle relative to the unmanned device. The map can be a high-precision map. Then, based on the current position of the unmanned device in the map and the position of each obstacle in the map, the road where the unmanned device is located, the initial planned path currently corresponding to the unmanned device, the position and outline of the unmanned device, and the position and outline of each obstacle are marked on the map.

[0078] In this way, the relative position between the initial planned path and each obstacle can be determined based on the initial planned path currently corresponding to the unmanned equipment and the current position of each obstacle. Figure 3 shown.

[0079] exist Figure 3 In the figure, the road where the unmanned device is located includes the unmanned device, obstacle C, obstacle D, and the initial planned path corresponding to the unmanned device. The initial planned path is located between obstacle C and the road boundary, that is, with the driving direction of the unmanned device as the front of the obstacle, the initial planned path is at least partially located on the left side of obstacle C.

[0080] When determining the expansion direction of the obstacle contour, the area around the obstacle may be first divided based on the edge on the obstacle contour to obtain multiple sub-areas, wherein one sub-area corresponds to one edge on the obstacle contour.

[0081] Specifically, for each edge on the outline of the obstacle, determine the line between the geometric center of the obstacle and the endpoint of the edge as the line corresponding to the edge. Then, according to the extension line of the line corresponding to each edge, divide the area around the obstacle to obtain multiple sub-areas. Specifically, the multiple areas enclosed by the intersection of the extension line of the line corresponding to each edge and the road boundary of the road area where the obstacle is located can be used as multiple sub-areas. Among them, the extension line of the line corresponding to each edge can refer to the extension line obtained by extending the line corresponding to each edge to the outside of the outline of the obstacle. Figure 4a shown.

[0082] exist Figure 4a In the figure, the outlines of obstacles C and D are taken as rectangles, and the extension lines corresponding to each side of the outlines of the obstacles are represented by dotted lines. The dotted lines corresponding to obstacle C divide the area around obstacle C into four sub-areas. Similarly, the dotted lines corresponding to obstacle D divide the area around obstacle D into four sub-areas.

[0083] In addition to dividing the area around the obstacle into multiple sub-areas based on the line between the geometric center of the obstacle and the endpoint of the edge, the area around the obstacle can also be divided into multiple sub-areas based on the extension line of the edge on the outline of the obstacle extending in a specified direction. Figure 4b shown.

[0084] exist Figure 4b In FIG. 5 , the edge on the outline of the obstacle is extended to one end to form an extension line. The extension line represented by the dotted line divides the area around the obstacle C and the obstacle D into four sub-areas respectively.

[0085] After the area around the obstacle is divided into multiple sub-areas, a sub-area that does not cover at least a portion of the initial planned path can be determined from the multiple sub-areas according to the position of the initial planned path and the area positions of the multiple sub-areas as a target sub-area, and the direction of the edge corresponding to the target sub-area away from the geometric center of the obstacle is used as the expansion direction of the obstacle contour.

[0086] S204: Determine, according to the system error, an expansion distance corresponding to the outline of the obstacle, wherein the larger the system error is, the larger the expansion distance is.

[0087] In order to expand the outline of the obstacle, it is necessary not only to determine the expansion direction of the obstacle outline, but also to determine the expansion distance of the obstacle outline. The expansion of the obstacle outline is to reduce the problem of large differences between the driving paths planned for each cycle caused by the system errors of the various systems of the driving system. Therefore, the expansion distance of the obstacle outline should be related to the system errors of each system. In addition, in order to avoid the expansion distance corresponding to the determined obstacle outline being too large or too small, the expansion distance corresponding to the obstacle outline can be determined based on the system error of at least one system in the driving system obtained. Among them, the larger the system error, the larger the expansion distance.

[0088] In the embodiments of the present specification, the expansion distance of the obstacle can be predicted by a regression model according to the system error involved in the driving system. The expansion distance of the obstacle can also be determined by a grid search method according to the system error involved in the driving system.

[0089] The first method is to use regression model.

[0090] Specifically, the system error involved in the driving system is input into a pre-trained prediction model, so that the prediction model predicts the expansion distance corresponding to the obstacle according to the regression relationship between the system error and the sample error used in the prediction model training process. The prediction model can be a regression model, and the regression model can include: a linear regression model, a Gaussian process regression model, etc.

[0091] Train the prediction model: Obtain the system error of the driving system used by the unmanned equipment in history as the sample error. Input each sample error into the prediction model to be trained so that the prediction model determines the regression relationship between each sample error according to the preset kernel function, and predicts the expansion distance to be optimized corresponding to the outline of the historical obstacle based on the regression relationship. Train the prediction model with the goal of minimizing the difference between the expansion distance to be optimized and the actual expansion distance corresponding to the outline of the historical obstacle.

[0092] In the process of training the prediction model, the system error corresponding to each driving system version in history can be used as the sample error. The preset kernel function can be used to determine the correlation between each sample error.

[0093] The second method: using grid search.

[0094] Specifically, according to the system error involved in the driving system, an initial expansion distance interval corresponding to the outline of the obstacle is determined, and the initial expansion distance interval is divided into equal intervals to obtain a plurality of sub-distance intervals.

[0095] Through the simulation system corresponding to the unmanned equipment, for each sub-distance interval, the expansion distance corresponding to the sub-distance interval is evaluated according to the driving path adjusted after the obstacle expands the sub-distance interval, and the evaluation result corresponding to the sub-distance interval is obtained. Among them, the simulation system refers to a system that simulates the driving environment of the unmanned equipment and simulates the unmanned equipment driving according to the driving path planned by the unmanned equipment. The adjusted driving path can refer to the planned path after the initial planned path is adjusted by the simulation system.

[0096] According to the evaluation result corresponding to each sub-distance interval, the expansion distance corresponding to the outline of the obstacle is selected from the initial expansion distance interval.

[0097] When evaluating the expansion distance corresponding to the sub-distance interval, the expansion distance corresponding to the sub-distance interval can be evaluated according to the difference between the driving path adjusted after the obstacle expands the expansion distance corresponding to the sub-distance interval and the initial planned path. The greater the difference between the driving path adjusted after the obstacle expands the expansion distance corresponding to the sub-distance interval and the initial planned path, the worse the evaluation result corresponding to the sub-distance interval.

[0098] When selecting the expansion distance corresponding to the outline of the obstacle from the initial expansion distance interval according to the evaluation result corresponding to each sub-distance interval, the sub-distance interval with the best evaluation result can be selected from multiple sub-distance intervals according to the evaluation result corresponding to each sub-distance interval as the target sub-distance interval. Then, the expansion distance corresponding to the outline of the obstacle is determined according to the target sub-distance interval.

[0099] Furthermore, the average distance of the target sub-distance interval can be determined as the expansion distance corresponding to the outline of the obstacle. The middle distance of the target sub-distance interval can also be determined as the expansion distance corresponding to the outline of the obstacle.

[0100] In addition, the simulation system in this specification can also evaluate other indicators of the adjusted driving path, where the other indicators may include: passability, safety, smoothness, etc.

[0101] Regarding safety: When the simulation system simulates the unmanned equipment driving along the adjusted driving path, the safety factor corresponding to the adjusted driving path is determined according to the distance between the obstacles around the unmanned equipment and the adjusted driving path. Among them, the greater the average distance between the unmanned equipment and the obstacles, the higher the safety factor.

[0102] Regarding passability: The simulation system simulates the unmanned equipment driving along the adjusted driving path, and judges whether the unmanned equipment can reach the destination according to the preset destination position. If it can reach the destination, it is determined that the adjusted driving path is passable.

[0103] Regarding smoothness: When the simulation system simulates the unmanned equipment driving along the adjusted driving path, it is determined whether the unmanned equipment has sudden braking behavior when driving along the adjusted driving path. If not, it is determined that the adjusted driving path is smooth.

[0104] In addition, the simulation system corresponding to the unmanned equipment can be applied to the unmanned equipment or to the server that controls the unmanned equipment.

[0105] In addition, after the prediction model is used to predict the expansion distance corresponding to the outline of the obstacle, in order to improve the accuracy of the expansion distance, the expansion distance corresponding to the outline of the obstacle can be predicted according to the prediction model, and the initial expansion distance interval corresponding to the outline of the obstacle can be determined. Then, the grid search method is used to re-determine the expansion distance corresponding to the outline of the obstacle.

[0106] It should be noted that there is no particular order in which step S202 and step S204 are executed.

[0107] S206: Expand the outline of the obstacle according to the expansion distance and the expansion direction to obtain the expanded outline of the obstacle.

[0108] In the embodiment of the present specification, the outline of the obstacle is expanded according to the expansion distance and the expansion direction to obtain the expanded outline of the obstacle.

[0109] Specifically, the edge of the obstacle contour in the expansion direction is expanded by the corresponding expansion distance in the direction away from the geometric center of the obstacle to obtain the expanded contour of the obstacle. Figure 5 shown.

[0110] exist Figure 5 In the example, the expansion distance is 2 cm. The edges on the outline of obstacle C are a, b, c, and d. The edges on the outline of obstacle D are e, f, g, and h. The edge on the outline of obstacle C in the expansion direction is c, and the edges on the outline of obstacle D in the expansion direction are g and h. Therefore, c, g, and h are expanded 2 cm in the direction away from the geometric center of the obstacle to obtain the expanded outlines of obstacles C and D. The expanded area is represented by a dotted line.

[0111] S208: adjusting the initial planned path according to the expanded outline of the obstacle to obtain an adjusted planned path, and controlling the unmanned equipment to travel according to the adjusted planned path.

[0112] In the embodiment of the present specification, the initial planned path is adjusted according to the expanded outline of the obstacle to obtain an adjusted planned path, and the unmanned equipment is controlled to travel according to the adjusted planned path. The difference between the adjusted planned path and the initial planned path is small. The adjustment may include: modifying part of the initial planned path or re-planning the path as the adjusted planned path.

[0113] The embodiments of this specification provide a schematic diagram of the initial planned path after adjustment, such as Figure 6 shown.

[0114] In addition, since in this specification, the current initial planned path is adjusted after the outline of the obstacle is expanded to obtain the adjusted planned path, when there is a large difference between the adjusted planned path and the initial planned path, it can be determined that the adjusted planned path is a more preferred planned path compared to the initial planned path. In the long run, the planned path of the unmanned equipment can be optimized.

[0115] Through the above Figure 2It can be seen from the method shown that this manual determines the expansion distance of the obstacle based on the system error of the driving system of the unmanned equipment, and at the same time, determines the expansion direction of the obstacle's outline based on the relative position between the obstacle's current initial planned path and the obstacle. Then, the outline of the obstacle is expanded according to the expansion direction and the expansion distance, and the initial planned path is adjusted based on the expanded outline. The unmanned equipment is controlled to travel according to the adjusted path. In this method, the outline of the obstacle is expanded in a certain direction and distance to reduce the passable area used to plan the driving path of the unmanned equipment. In this way, the amplitude of adjusting the initial planned path can be reduced, thereby reducing the frequency of changing the driving path of the unmanned equipment.

[0116] Further, in Figure 2 In step S204 shown, taking the prediction model as a Gaussian process regression model and the system error including perception error, positioning error, prediction error, and control error as an example, the Gaussian process regression model is trained.

[0117] When obtaining the system error corresponding to the historical version of the driving system used by the unmanned equipment as the sample error, it is assumed that the system error of the historical version of the driving system is: X = [x 1 , x 2 , …, x i ] T , x i =[σ i1 , σ i2 , σ i3 , σ i4 ]. Among them, σ i1 , σ i2 , σ i3 , σ i4 are the perception error, positioning error, prediction error, and control error of the driving system in the i-th historical version, respectively. 1 , x 2 , …, x i ] T The T in stands for transpose.

[0118] The actual expansion distance of the contour corresponding to the historical obstacle contour is Y = [y 1 ,y 2 , …, y i ] T ,y i =d i .d i is the actual expansion distance.

[0119] Among them, the kernel function of the Gaussian process regression model is the square exponential kernel, which is Among them, x and x′ represent the system errors corresponding to any two historical versions of the driving system, τ represents a hyperparameter, τ satisfies greater than zero, and the value of τ can be customized. The formula for the predicted expansion distance is: y = f(x i )+ξ i Among them, ξ i ~N(0,σ 2 ) is the noise matrix, f(x i ) represents the regression relationship function corresponding to the Gaussian regression model after training.

[0120] After the prediction model is trained, the system error of the current driving system is obtained. Assume that X* = {x *}, x * =[σ *1 , σ *2 , σ *3 , σ *4 ]. Among them, σ *1 , σ *2 , σ *3 , σ *4 They are the perception error, positioning error, prediction error and control error of the current driving system. * .d * is the expansion distance to be predicted.

[0121] The expansion distance to be predicted follows the Gaussian distribution Y * |x 1 , x * ,y 1 ~N(μ*,∑*). μ* is the mean, ∑* is the covariance matrix. Among them, μ*=K SE (X, X*)(K SE (X, X)+σ 2 I) -1 y.∑*=K SE (X*, X*)-K SE (X*,X)(K SE (X, X)+σ 2 I) -1 K SE (X, X*). I represents the identity matrix. Finally, when determining the dilation distance, the maximum likelihood estimate of the Gaussian distribution is obtained.

[0122] The above is a control method for unmanned equipment provided in the embodiments of this specification. Based on the same idea, this specification also provides corresponding devices, storage media and electronic devices.

[0123] Figure 7 A schematic diagram of a structure of an unmanned equipment control device provided in an embodiment of this specification, the device comprising:

[0124] An acquisition module 701 is used to acquire an initial planned path currently corresponding to the unmanned device and a system error of a driving system used by the unmanned device;

[0125] A first determination module 702 is used to determine the expansion direction of the outline of the obstacle according to the relative position between the initial planned path and the obstacle within the preset range of the unmanned equipment;

[0126] A second determination module 703 determines, according to the system error, an expansion distance corresponding to the outline of the obstacle, wherein the larger the system error is, the larger the expansion distance is;

[0127] An expansion module 704, configured to expand the outline of the obstacle according to the expansion distance and the expansion direction to obtain the expanded outline of the obstacle;

[0128] The control module 705 is used to adjust the initial planned path according to the expanded outline of the obstacle to obtain an adjusted planned path, and control the unmanned equipment to travel according to the adjusted planned path.

[0129] Optionally, the driving system includes: at least one of a perception system, a positioning system, a prediction system, and a control system;

[0130] The system error of the perception system is a perception error, and the perception error refers to the error between the position of the obstacle perceived by the perception system and the actual position of the obstacle;

[0131] The system error of the positioning system is the positioning error, and the positioning error refers to the error between the positioning position of the unmanned equipment determined by the positioning system and the actual position of the unmanned equipment;

[0132] The system error of the prediction system is the prediction error, and the prediction error refers to the error between the trajectory position of the obstacle predicted by the prediction system and the actual position of the obstacle;

[0133] The system error of the control system is a control error, and the control error refers to the error between the target position that the unmanned equipment should reach under the control system and the positioning position of the unmanned equipment.

[0134] Optionally, the first determination module 702 can be specifically used to divide the area around the obstacle based on the edge on the contour of the obstacle to obtain multiple sub-areas; wherein one sub-area corresponds to one edge on the contour; according to the position of the initial planned path and the area positions of the multiple sub-areas, determine from the multiple sub-areas a sub-area that does not cover at least part of the initial planned path as a target sub-area; and use the direction of the edge corresponding to the target sub-area away from the geometric center of the obstacle as the expansion direction of the contour of the obstacle.

[0135] Optionally, the first determination module 702 can be specifically used to determine, for each edge on the outline of the obstacle, a line between the geometric center of the obstacle and the endpoint of the edge as the line corresponding to the edge; and to use multiple areas enclosed by the intersection of an extension line of the line corresponding to each edge and a road boundary of the road area where the obstacle is located as multiple sub-areas.

[0136] Optionally, the second determination module 703 can be specifically used to input the system error into a pre-trained prediction model so that the prediction model predicts the expansion distance corresponding to the obstacle based on the regression relationship between the system error and the sample error used in the prediction model training process.

[0137] Optionally, before inputting the system error into a pre-trained prediction model, the second determination module 703 can also be used to obtain the system error of the driving system used by the unmanned equipment in history as a sample error; input each sample error into the prediction model to be trained, so that the prediction model predicts the expansion distance to be optimized corresponding to the contour of the historical obstacle based on the regression relationship between each sample error; and train the prediction model with the goal of minimizing the difference between the expansion distance to be optimized and the actual expansion distance corresponding to the contour of the historical obstacle.

[0138] Optionally, the second determination module 703 can be specifically used to determine, based on the system error, an initial expansion distance interval corresponding to the outline of the obstacle; divide the initial expansion distance interval into equal intervals to obtain a plurality of sub-distance intervals; through a simulation system corresponding to the unmanned equipment, for each sub-distance interval, evaluate the expansion distance corresponding to the sub-distance interval according to the driving path adjusted after the obstacle expands the sub-distance interval, and obtain an evaluation result corresponding to the sub-distance interval; and select the expansion distance corresponding to the outline of the obstacle from the initial expansion distance interval based on the evaluation result corresponding to each sub-distance interval.

[0139] This specification also provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, it can be used to perform the above Figure 2 A control method for unmanned equipment is provided.

[0140] The embodiments of this specification also provide Figure 8 The structural diagram of the electronic device shown in FIG. Figure 8 At the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include other hardware required for the business. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to achieve the above Figure 2 The control method of the unmanned equipment.

[0141] Of course, in addition to software implementation, this specification does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc., that is to say, the executor of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0142] In the 1990s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with the development of technology, many improvements to the method flow today can be regarded as direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0143] The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.

[0144] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0145] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0146] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0147] This specification is described with reference to the flowchart and / or block diagram of the method, device (system), and computer program product according to the embodiment of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the process and / or box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one process or multiple processes in the flowchart and / or one box or multiple boxes in the block diagram.

[0148] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0149] These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0150] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0151] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0152] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0153] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0154] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0155] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0156] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0157] The above description is only an embodiment of the present specification and is not intended to limit the present specification. For those skilled in the art, the present specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification shall be included in the scope of the claims of the present specification.

Claims

1. A control method for unmanned equipment, characterized in that: include: Obtaining an initial planned path currently corresponding to the unmanned device and a system error of a driving system used by the unmanned device; Determining the expansion direction of the outline of the obstacle according to the relative position between the initial planned path and the obstacle within the preset range of the unmanned equipment; Determine, according to the system error, an expansion distance corresponding to the outline of the obstacle, wherein the larger the system error is, the larger the expansion distance is; Expanding the outline of the obstacle according to the expansion distance and the expansion direction to obtain the expanded outline of the obstacle; According to the expanded outline of the obstacle, the initial planned path is adjusted to obtain an adjusted planned path, and the unmanned equipment is controlled to travel according to the adjusted planned path; The step of determining the expansion direction of the obstacle contour according to the relative position between the initial planned path and the obstacle within the preset range of the unmanned equipment includes: Based on the edges on the outline of the obstacle, the area around the obstacle is divided to obtain a plurality of sub-areas; Wherein, a sub-region corresponds to an edge on the contour; According to the position of the initial planned path and the area positions of the multiple sub-areas, determining a sub-area that does not cover at least a portion of the initial planned path from the multiple sub-areas as a target sub-area; Taking the direction in which the edge corresponding to the target sub-region is away from the geometric center of the obstacle as the expansion direction of the outline of the obstacle; The area around the obstacle is divided based on the edge on the outline of the obstacle to obtain multiple sub-areas, including: For each edge on the outline of the obstacle, determine a line between the geometric center of the obstacle and the endpoint of the edge as the line corresponding to the edge; A plurality of regions enclosed by the intersection of the extended line of the connecting line corresponding to each edge and the road boundary of the road region where the obstacle is located are used as a plurality of sub-regions.

2. The method according to claim 1, characterized in that The driving system includes: at least one of a perception system, a positioning system, a prediction system, and a control system; The system error of the perception system is a perception error, and the perception error refers to the error between the position of the obstacle perceived by the perception system and the actual position of the obstacle; The system error of the positioning system is the positioning error, and the positioning error refers to the error between the positioning position of the unmanned equipment determined by the positioning system and the actual position of the unmanned equipment; The system error of the prediction system is the prediction error, and the prediction error refers to the error between the trajectory position of the obstacle predicted by the prediction system and the actual position of the obstacle; The system error of the control system is a control error, and the control error refers to the error between the target position that the unmanned equipment should reach under the control system and the positioning position of the unmanned equipment.

3. The method according to claim 1, characterized in that Determining, according to the system error, an expansion distance corresponding to the outline of the obstacle, comprising: The system error is input into a pre-trained prediction model, so that the prediction model predicts the expansion distance corresponding to the obstacle according to the regression relationship between the system error and the sample error used in the prediction model training process.

4. The method according to claim 3, characterized in that Before inputting the system error into the pre-trained prediction model, the method further includes: Obtain the system error of the driving system used by the unmanned equipment in history as the sample error; Inputting each sample error into the prediction model to be trained, so that the prediction model predicts the expansion distance to be optimized corresponding to the contour of the historical obstacle according to the regression relationship between each sample error; The prediction model is trained with the goal of minimizing the difference between the expansion distance to be optimized and the actual expansion distance corresponding to the contour of the historical obstacle.

5. The method according to claim 1, characterized in that Determining, according to the system error, an expansion distance corresponding to the outline of the obstacle, comprising: Determining an initial expansion distance interval corresponding to the outline of the obstacle according to the system error; Dividing the initial expansion distance interval into equal intervals to obtain a plurality of sub-distance intervals; By means of the simulation system corresponding to the unmanned equipment, for each sub-distance interval, the expanded distance corresponding to the sub-distance interval is evaluated according to the driving path adjusted after the obstacle expands the expanded distance corresponding to the sub-distance interval, so as to obtain an evaluation result corresponding to the sub-distance interval; According to the evaluation result corresponding to each sub-distance interval, the expansion distance corresponding to the outline of the obstacle is selected from the initial expansion distance interval.

6. An unmanned equipment control device, characterized in that: include: An acquisition module is used to acquire an initial planned path currently corresponding to the unmanned equipment and a system error of a driving system used by the unmanned equipment; A first determination module, configured to determine an expansion direction of a contour of the obstacle according to a relative position between the initial planned path and an obstacle within a preset range of the unmanned equipment; A second determination module determines, according to the system error, an expansion distance corresponding to the outline of the obstacle, wherein the larger the system error is, the larger the expansion distance is; An expansion module, used to expand the outline of the obstacle according to the expansion distance and the expansion direction to obtain the expanded outline of the obstacle; A control module, configured to adjust the initial planned path according to the expanded outline of the obstacle to obtain an adjusted planned path, and control the unmanned equipment to travel according to the adjusted planned path; The step of determining the expansion direction of the obstacle contour according to the relative position between the initial planned path and the obstacle within the preset range of the unmanned equipment includes: Based on the edges on the outline of the obstacle, the area around the obstacle is divided to obtain a plurality of sub-areas; Wherein, a sub-region corresponds to an edge on the contour; According to the position of the initial planned path and the area positions of the multiple sub-areas, determining a sub-area that does not cover at least a portion of the initial planned path from the multiple sub-areas as a target sub-area; Taking the direction in which the edge corresponding to the target sub-region is away from the geometric center of the obstacle as the expansion direction of the outline of the obstacle; The area around the obstacle is divided based on the edge on the outline of the obstacle to obtain multiple sub-areas, including: For each edge on the outline of the obstacle, determine a line between the geometric center of the obstacle and the endpoint of the edge as the line corresponding to the edge; A plurality of regions enclosed by the intersection of the extended line of the connecting line corresponding to each edge and the road boundary of the road region where the obstacle is located are used as a plurality of sub-regions.

7. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method described in any one of claims 1 to 5 is implemented.

Images