Target object searching method and device, computer equipment and storage medium

Abstract

The invention relates to a target object searching method and device, computer equipment and a storage medium. When it is determined that the unmanned vehicle does not search the target object according to the current environment information of the unmanned vehicle, obtaining current pose information of the unmanned vehicle; determining boundary points of an unsearched area according to the current environment information, the searched area and the current pose information; determining a target point according to a searched range proportion corresponding to the boundary point, the course angle difference value and the position spacing; and further controlling the unmanned vehicle to search the target object in the process of driving from the current position to the target point. According to the scheme, the target point is determined according to the searched range proportion, the course angle difference value and the position distance corresponding to the boundary point, the time for the unmanned vehicle to arrive at the next target point for searching can be saved, and therefore the target object searching efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent driving, and particularly relates to a method, device, computer device, and storage medium for searching for a target object. Background Art

[0002] With the rapid development of science and technology, the intelligent driving technology of vehicles has gradually matured. The intelligent driving technology of driverless vehicles can be applied to complex and dangerous unknown environments such as the wild to autonomously search for the environment, search for people, search for supplies, etc., which can effectively reduce the risk of search tasks and improve search efficiency.

[0003] In traditional technologies, a frontier search tree is usually established and extracted based on environmental data and vehicle pose data, and the frontier search tree is traversed based on a depth-first algorithm search strategy, and then a driving trajectory is planned and the vehicle is controlled to drive to complete the search of an unknown environment.

[0004] However, the depth-first algorithm has problems such as possible entrapment in invalid branches, a large number of backtracking operations, and lack of global optimality, resulting in low target object search efficiency. Summary of the Invention

[0005] Based on this, in view of the above technical problems, it is necessary to provide a method, device, computer device, and storage medium for searching for a target object, which can improve the search efficiency of the target object.

[0006] In a first aspect, the present application provides a method for searching for a target object, including:

[0007] When it is determined according to the current environment information of the driverless vehicle that the target object has not been searched, obtain the current pose information of the driverless vehicle;

[0008] Determine the boundary points of the unsearched area according to the current environment information, the searched area, and the current pose information;

[0009] Determine the target point according to the searched range ratio corresponding to the boundary point, the heading angle difference, and the position distance; the searched range ratio is the ratio between the searched area and the preset area within the preset area corresponding to the boundary point; the heading angle difference is the difference between the heading angle of the current position in the current pose information of the driverless vehicle and the target angle, and the target angle is the angle between the direction from the current position to the boundary point and the reference direction; the position distance is the spatial distance between the current position and the boundary point;

[0010] Control the driverless vehicle to search for the target object during the process of driving from the current position to the target point.

[0011] In one of the embodiments, the number of boundary points is at least two; determining the target point according to the searched range ratio corresponding to the boundary point, the heading angle difference, and the position distance includes:

[0012] For each boundary point, perform a weighted operation on the ratio of the searched range corresponding to the boundary point, the heading angle difference, and the position spacing to obtain a cost function value;

[0013] Take the boundary point corresponding to the minimum cost function value among all boundary points as the target point.

[0014] In one embodiment, determining the boundary points of the unsearched area according to the current environmental information, the searched area, and the current pose information includes:

[0015] Establish a global map of the search area according to the current environmental information, the map corresponding to the searched area, and the current pose information; the search area includes the searched area and the current search area;

[0016] During the process of expanding the area starting from the current position, determine the boundary points of the unsearched area according to the global map and the expanded area.

[0017] In one embodiment, the current environmental information includes the image information and the laser point cloud information of the current search area. Establishing a global map of the search area according to the current environmental information, the map corresponding to the searched area, and the current pose information includes:

[0018] Obtain the image information and the laser point cloud information of the unmanned vehicle in the current search area;

[0019] Generate a spatial occupancy grid map of the unmanned vehicle in the current search area according to the image information and the laser point cloud information of the unmanned vehicle in the current search area, and the current pose information;

[0020] Stitch the spatial occupancy grid map of the unmanned vehicle in the current search area with the map corresponding to the searched area to obtain the global map of the search area.

[0021] In one embodiment, during the process of expanding the area starting from the current position, determining the boundary points of the unsearched area according to the global map and the expanded area includes:

[0022] Perform grid division on the global map and the expanded area;

[0023] During the process of expanding the area starting from the current position, if it is determined according to the global map and the expanded area that the grid before expansion belongs to the detected area, while the grid after expansion belongs to the undetected area, then take the grid before expansion as the boundary point.

[0024] In one embodiment, controlling the unmanned vehicle to search for the target object during the process of driving from the current position to the target point includes:

[0025] Generate a driving path from the current position to the target point subject to path generation conditions; the path generation conditions are that the driving path is the shortest in distance and there are no obstacles in the driving path.

[0026] During the process of controlling the unmanned vehicle to drive along the driving path, search for the target object.

[0027] In a second aspect, the present application also provides a target object search device, including:

[0028] An acquisition module, configured to acquire the current pose information of the unmanned vehicle when it is determined according to the current environmental information of the unmanned vehicle that the unmanned vehicle has not searched for the target object.

[0029] A first determination module, configured to determine the boundary points of the unsearched area according to the current environmental information, the searched area, and the current pose information.

[0030] A second determination module, configured to determine the target point according to the searched range ratio corresponding to the boundary point, the heading angle difference, and the position spacing; the searched range ratio is the ratio between the searched area and the preset area within the preset area corresponding to the boundary point; the heading angle difference is the difference between the heading angle of the unmanned vehicle at the current position and the target angle, and the target angle is the angle between the direction from the current position to the boundary point and the reference direction; the position spacing is the spatial distance between the current position and the boundary point.

[0031] A search module, configured to control the unmanned vehicle to search for the target object during the process of driving from the current position to the target point.

[0032] In a third aspect, the present application also provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0033] When it is determined according to the current environmental information of the unmanned vehicle that the unmanned vehicle has not searched for the target object, acquire the current pose information of the unmanned vehicle.

[0034] Determine the boundary points of the unsearched area according to the current environmental information, the searched area, and the current pose information.

[0035] Determine a target point based on the searched range ratio, heading angle difference, and position distance corresponding to the boundary point; the searched range ratio is the ratio between the searched area and the preset area within the preset area corresponding to the boundary point; the heading angle difference is the difference between the heading angle of the current position of the unmanned vehicle in the current pose information and the target angle, and the target angle is the angle between the direction from the current position to the boundary point and the reference direction; the position distance is the spatial distance between the current position and the boundary point.

[0036] Control the unmanned vehicle to search for the target object during the process of driving from the current position to the target point.

[0037] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0038] In the case where it is determined according to the current environmental information of the unmanned vehicle that the unmanned vehicle has not searched for the target object, obtain the current pose information of the unmanned vehicle.

[0039] Determine the boundary points of the unsearched area according to the current environmental information, the searched area, and the current pose information.

[0040] Determine a target point based on the searched range ratio, heading angle difference, and position distance corresponding to the boundary point; the searched range ratio is the ratio between the searched area and the preset area within the preset area corresponding to the boundary point; the heading angle difference is the difference between the heading angle of the current position of the unmanned vehicle in the current pose information and the target angle, and the target angle is the angle between the direction from the current position to the boundary point and the reference direction; the position distance is the spatial distance between the current position and the boundary point.

[0041] Control the unmanned vehicle to search for the target object during the process of driving from the current position to the target point.

[0042] In a fifth aspect, the present application further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0043] In the case where it is determined according to the current environmental information of the unmanned vehicle that the unmanned vehicle has not searched for the target object, obtain the current pose information of the unmanned vehicle.

[0044] Determine the boundary points of the unsearched area according to the current environmental information, the searched area, and the current pose information.

[0045] Determine a target point based on the searched range ratio, heading angle difference, and position spacing corresponding to the boundary point; the searched range ratio is the ratio between the searched area and the preset area within the preset area corresponding to the boundary point; the heading angle difference is the difference between the heading angle of the current position of the unmanned vehicle in the current pose information and the target angle, where the target angle is the angle between the direction from the current position to the boundary point and the reference direction; the position spacing is the spatial distance between the current position and the boundary point.

[0046] Control the unmanned vehicle to search for the target object during the process of driving from the current position to the target point.

[0047] In the above target object search method, device, computer device, and storage medium, when it is determined according to the current environmental information of the unmanned vehicle that the target object has not been searched, obtain the current pose information of the unmanned vehicle; and determine the boundary points of the unsearched area according to the current environmental information, the searched area, and the current pose information; determine the target point according to the searched range ratio, heading angle difference, and position spacing corresponding to the boundary point; and then control the unmanned vehicle to search for the target object during the process of driving from the current position to the target point. In the above solution, the target point is determined according to the searched range ratio, heading angle difference, and position spacing corresponding to the boundary point, so that the determined target point is, relative to the unmanned vehicle, the boundary point with the smallest cost for the unmanned vehicle to reach the boundary point from the current position. In this way, the time for the unmanned vehicle to reach the next target point for searching can be saved, thereby improving the search efficiency of the target object. Description of the Drawings

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0049] Figure 1 It is a schematic flowchart of the target object search method in an embodiment;

[0050] Figure 2 It is a schematic flowchart of determining the target point in an embodiment;

[0051] Figure 3 It is a schematic flowchart of determining the boundary points of the unsearched area in an embodiment;

[0052] Figure 4 It is a schematic flowchart of establishing a global map of the search area in an embodiment;

[0053] Figure 5 Schematic diagram of the process for determining the boundary points of the unsearched area in another embodiment;

[0054] Figure 6 Schematic diagram of the process for searching for the target object in one embodiment;

[0055] Figure 7 Schematic diagram of the process for the target object search method in another embodiment;

[0056] Figure 8 Structural block diagram of the target object search device in one embodiment;

[0057] Figure 9 Internal structure diagram of a computer device in one embodiment. Detailed implementation manners

[0058] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0059] The target object search method provided by the embodiments of the present application can be applied to the application scenario of controlling an autonomous vehicle to search for a target object. This method can be applied to a server or the intelligent driving controller of a vehicle. Among them, the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0060] In an exemplary embodiment, as Figure 1 shown, a target object search method is provided. Taking the application of this method to the intelligent driving controller of a vehicle as an example, the method includes the following steps:

[0061] S101, when it is determined according to the current environmental information of the autonomous vehicle that the autonomous vehicle has not searched for the target object, obtain the current pose information of the autonomous vehicle.

[0062] Exemplarily, the driverless vehicle can autonomously search for target objects in complex and dangerous unknown environments such as the wild through autonomous driving technology. The target objects can be people, objects, the environment, etc. The current environmental information of the driverless vehicle can be obtained through sensors such as cameras and lidar mounted on the driverless vehicle. Among them, the current environmental information can reflect the object distribution and spatial structure of the environment around the driverless vehicle. In the case where it is determined that the driverless vehicle has not searched for the target object based on the current environmental information of the driverless vehicle, the current pose information of the driverless vehicle itself can be collected in real time through sensors such as satellite positioning and inertial navigation. Among them, the current pose information includes current position information and attitude information. The current position information is used to reflect the coordinates of the driverless vehicle in space, and the attitude information is used to reflect the states such as the orientation of the driverless vehicle.

[0063] S102. Determine the boundary points of the unsearched area according to the current environmental information, the searched area, and the current pose information.

[0064] Exemplarily, the current search area of the driverless vehicle can be determined according to the current pose information of the driverless vehicle and the current environmental information of the driverless vehicle, and the unsearched area can be determined according to the searched area and the current search area; and the points on the boundary of the area that has not been searched are used as the boundary points of the unsearched area.

[0065] S103. Determine the target point according to the searched range ratio corresponding to the boundary point, the heading angle difference, and the position spacing.

[0066] Among them, the searched range ratio can be understood as the ratio between the searched area and the preset area within the preset area corresponding to the boundary point; the preset area corresponding to the boundary point can be preset, for example, it can be set as a circular area or a rectangular area with a preset size centered on the boundary point; it can also be the entire area to be searched determined in advance.

[0067] The heading angle difference can be understood as the difference or the absolute value of the difference between the heading angle of the driverless vehicle at the current position and the target angle. Among them, the target angle is the angle between the direction from the current position to the boundary point and the reference direction; the reference direction can be a preset direction, for example, it can be the due south or due north direction. The heading angle of the driverless vehicle at the current position is the angle between the heading of the driverless vehicle at the current position and the reference direction.

[0068] The position spacing can be understood as the spatial distance between the current position and the boundary point.

[0069] Exemplarily, a target point can be determined among each boundary point. The target point can be understood as the end point for the next target object search by the driverless vehicle. Thus, for each boundary point, based on the cost function, according to the ratio of the searched range corresponding to the boundary point, the heading angle difference, and the position spacing, the cost function value corresponding to each boundary point can be determined. Furthermore, the boundary point with the minimum cost function value is taken as the target point. In this way, the time and resources consumed by the driverless vehicle from the current position to the target point can be saved.

[0070] Exemplarily, the sum of the ratio of the searched range corresponding to the boundary point, the heading angle difference, and the position spacing can be directly used as the cost function value, or the ratio of the searched range corresponding to the boundary point, the heading angle difference, and the position spacing can be subjected to a weight operation, and the sum of the ratio of the searched range, the heading angle difference, and the position spacing after the weight operation is used as the cost function value.

[0071] S104. Control the driverless vehicle to search for the target object during the process of driving from the current position to the target point.

[0072] Furthermore, the driverless vehicle can be controlled to move forward from the current position to the target point. For example, a driving path can be generated in advance, and then the driverless vehicle can be controlled to move forward from the current position to the target point according to the driving path; or the driving direction and speed and other data can be determined according to the surrounding environment information of the real-time position to control the driverless vehicle to move forward from the current position to the target point.

[0073] During the process of controlling the driverless vehicle to drive from the current position to the target point, control the driverless vehicle to search for the target object. Exemplarily, the target object can be identified according to the target object recognition algorithm or software in the driverless vehicle, where the target object recognition algorithm or software is pre-trained according to the sample object and the sample label.

[0074] If the target object is searched, the task ends and the driverless vehicle stops searching. If the target object is not searched, the driverless vehicle repeats the steps of S101 - S104 until the target object is searched or the entire area to be searched is traversed.

[0075] In the above object search method, when it is determined according to the current environmental information of the driverless vehicle that the object has not been searched, the current pose information of the driverless vehicle is obtained; and according to the current environmental information of the driverless vehicle, the searched area, and the current pose information, the boundary points of the unsearched area are determined; according to the searched range ratio, the heading angle difference, and the position spacing corresponding to the boundary points, the target point is determined; and then, during the process of controlling the driverless vehicle to travel from the current position to the target point, the object is searched. In the above solution, according to the searched range ratio, the heading angle difference, and the position spacing corresponding to the boundary points, the target point is determined, so that the determined target point is, relative to the driverless vehicle, the boundary point with the smallest cost for the driverless vehicle to reach the boundary point from the current position. In this way, the time for the driverless vehicle to reach the next target point for searching can be saved, thereby improving the search efficiency of the object.

[0076] In some optional implementation manners, the number of boundary points is at least two. A weighted operation can be performed on the searched range ratio, the heading angle difference, and the position spacing corresponding to the boundary points to obtain a cost function value, and then the target point is determined according to the cost function value.

[0077] Based on this, referring to Figure 2 , Figure 2 a flowchart for determining the target point is provided, which specifically includes the following steps:

[0078] S201. For each boundary point, a weighted operation is performed on the searched range ratio, the heading angle difference, and the position spacing corresponding to the boundary point to obtain a cost function value.

[0079] Exemplarily, a real vehicle test can be performed on the driverless vehicle to determine the weight coefficients corresponding to the searched range ratio, the heading angle difference, and the position spacing respectively. Then, the searched range ratio, the heading angle difference, and the position spacing corresponding to the boundary point are multiplied by the corresponding weight coefficients respectively, and the sum of the products is used as the cost function value.

[0080] S202. The boundary point corresponding to the minimum cost function value among the boundary points is used as the target point.

[0081] Exemplarily, each cost function value is used to represent the cost that the driverless vehicle needs to pay to reach each boundary point from the current position. The smaller the cost function value, the smaller the cost that needs to be paid. The smaller the cost paid, the faster the driverless vehicle can reach the boundary point with less resources consumed. Therefore, the boundary point corresponding to the minimum cost function value can be used as the target point, so that the driverless vehicle can reach the target point faster and with less resources consumed, in order to search for the object more efficiently and with less resources spent.

[0082] In the embodiments of the present application, by performing a weighted operation on the ratio of the searched range corresponding to the boundary point, the heading angle difference, and the position spacing, a cost function value is obtained, and then the boundary point corresponding to the minimum cost function value is used as the target point, achieving a more efficient and resource-saving search for the target object, thereby improving the search efficiency of the target object.

[0083] In some alternative implementation manners, the boundary points of the unsearched area can be determined by establishing a map of the search area.

[0084] Exemplarily, referring to Figure 3 , Figure 3 a flow schematic diagram for determining the boundary points of the unsearched area is provided, which specifically includes the following steps:

[0085] S301. Establish a global map of the search area according to the current environmental information, the map corresponding to the searched area, and the current pose information.

[0086] Exemplarily, the search area may include the searched area and the current search area. The map of the current search area can be established according to the current pose information of the unmanned vehicle and the current environmental information of the unmanned vehicle; and the map of the searched area can be established according to the environmental information corresponding to the searched area; and then the map of the current search area and the map of the searched area are stitched together to obtain the global map of the search area.

[0087] S302. During the process of expanding the area starting from the current position, determine the boundary points of the unsearched area according to the global map and the expanded area.

[0088] Exemplarily, the area can be expanded starting from the current position with a preset step length. During the process of expanding the area starting from the current position, the boundary points of the unsearched area can be determined according to the global map and the expanded area. For example, the global map can be understood as the map corresponding to the area that has been searched, and thus the points at the junction of the global map and the expanded area can be used as the boundary points of the unsearched area.

[0089] In the embodiments of the present application, by establishing a global map corresponding to the area that has been searched, expanding the area, and determining the boundary points of the unsearched area according to the global map and the expanded area, it can be ensured that the determined boundary points of the unsearched area do not include the boundary points that have been searched, avoiding the problem of repeated search for the area, and thereby improving the search efficiency.

[0090] In some alternative implementations, the current environmental information in the above embodiments may include image information and laser point cloud information of the current search area. Thus, a global map of the search area can be established based on the image information, the laser point cloud information, the map corresponding to the searched area, and the current pose information of the unmanned vehicle.

[0091] Based on this, referring to Figure 4 , Figure 4 a schematic flowchart of a process for establishing a global map of a search area is provided, which specifically includes the following steps:

[0092] S401, Obtain the image information and laser point cloud information of the unmanned vehicle in the current search area.

[0093] Exemplarily, the image information of the unmanned vehicle in the current search area can be obtained through a camera mounted on the unmanned vehicle, and the laser point cloud information of the unmanned vehicle in the current search area can be obtained through a laser sensor mounted on the unmanned vehicle. Among them, the image information and laser point cloud information of the unmanned vehicle in the current search area can be used to reflect the object distribution and spatial structure of the surrounding environment of the unmanned vehicle at the current position.

[0094] S402, Generate a spatial occupancy grid map of the unmanned vehicle in the current search area according to the image information, the laser point cloud information, and the current pose information of the unmanned vehicle in the current search area.

[0095] Exemplarily, the image information, the laser point cloud information, and the current pose information of the unmanned vehicle in the current search area can be fused to generate a spatial occupancy grid map of the unmanned vehicle in the current search area. Among them, the spatial occupancy grid map of the unmanned vehicle in the current search area can divide the space into grid cells one by one, and each grid cell determines whether it is occupied by an object according to the environmental information, so as to visually present the spatial occupancy situation of the local area around the unmanned vehicle in the form of a grid.

[0096] S403, Stitch the spatial occupancy grid map of the unmanned vehicle in the current search area with the map corresponding to the searched area to obtain the global map of the search area.

[0097] Exemplarily, the map corresponding to the searched area can also be generated according to the generation method of the spatial occupancy grid map of the unmanned vehicle in the current search area described above. Furthermore, the spatial occupancy grid map of the unmanned vehicle in the current search area can be stitched with the map corresponding to the searched area. During the stitching process, the current position information can be used as a reference benchmark to accurately fuse the spatial occupancy grid map of the unmanned vehicle in the current search area into the map framework corresponding to the searched area, so that the newly detected unknown area part is seamlessly connected to the detected area, and the global map of the search area is obtained.

[0098] In the embodiments of the present application, by fusing the environmental information in the form of images and laser point clouds around the driverless vehicle with the map corresponding to the searched area and the current pose information, a global map of the searched area is obtained, so that the global map of the searched area can more accurately reflect the environment around the driverless vehicle, facilitating subsequent target search.

[0099] In some alternative implementation manners, based on Figure 4 the embodiment shown, the area can be expanded in a preset number of grids, and then the boundary points of the unsearched area can be determined.

[0100] Based on this, referring to Figure 5 , Figure 5 a schematic flow chart of another method for determining the boundary points of the unsearched area is provided, which specifically includes the following steps:

[0101] S501, perform grid division on the global map and the expanded area.

[0102] Exemplarily, the global map can be a spatial occupancy grid map, in which case there is no need to perform grid division on the global map again. If the global map is not a spatial occupancy grid map, grid division can be performed on the global map. And the expanded area can be divided into grids, so that the area can be expanded in units of grids to more accurately determine the boundary points of the unsearched area.

[0103] S502, during the process of expanding the area starting from the current position, if it is determined according to the global map and the expanded area that the grid before expansion belongs to the detected area and the grid after expansion belongs to the undetected area, then the grid before expansion is used as the boundary point.

[0104] Furthermore, the area can be expanded in a preset number of grids. For example, the area can be expanded in 8 grids. During the process of expanding the area starting from the current position, the boundary points can be determined according to the attributes of the grids. For example, if it is determined according to the global map and the expanded area that the grid before expansion belongs to the detected area and the grid after expansion belongs to the undetected area, then the grid before expansion is used as the boundary point.

[0105] In the embodiments of the present application, by performing grid division on the global map and the expanded area, it is convenient to expand the area in units of grids, which can make the granularity of the determined boundary points smaller, and thus avoid repeated search of the area to improve the efficiency of target search.

[0106] In some alternative implementation manners, during the process of controlling the driverless vehicle to travel from the current position to the target point, a travel path can be generated first, and the driverless vehicle can be controlled to travel according to the travel path to increase the travel speed, thereby improving the efficiency of target search.

[0107] Based on this, referring to Figure 6 , Figure 6 a flowchart of a process for searching for a target object is provided, which specifically includes the following steps:

[0108] S601, generate a driving path from the current position to the target point with the path generation conditions as constraints.

[0109] Exemplarily, the path generation conditions are used to guide the generation of the path. For example, the path generation conditions can be that the distance of the driving path is the shortest and there are no obstacles in the driving path, so that the determined driving path is optimal and will not collide with obstacles.

[0110] Exemplarily, the Rapidly-exploring Random Tree (RRT) algorithm can be used to generate an optimal path. For example, the algorithm can start from the current position of the driverless vehicle, randomly sample nodes in the search space, and gradually grow towards the target point by continuously expanding the tree structure. During the expansion process, the algorithm will consider the obstacle information in the global map to avoid the generated path passing through obstacles. After multiple iterations and searches, a feasible path from the current position to the target point is finally found, and the path is optimized according to the path generation conditions (such as the shortest path length, the least driving time, etc.) to obtain the optimal driving path.

[0111] S602, search for the target object while controlling the driverless vehicle to drive along the driving path.

[0112] Furthermore, while controlling the driverless vehicle to drive along the driving path, the target object can be searched. For example, the driving path can be parsed first, that is, the generated optimal driving path is parsed, and the path information is converted into a series of discrete points and corresponding driving parameters (such as speed, direction, etc.) so that the control system of the driverless vehicle can understand and execute.

[0113] Furthermore, a trajectory tracking algorithm based on Model Predictive Control (MPC) can be used to establish a model. For example, first establish the dynamic and kinematic models of the driverless vehicle, and this model describes the changes in the motion state of the driverless vehicle under different inputs (such as throttle, steering, etc.).

[0114] Furthermore, predict the future state of the driverless vehicle. For example, according to the current state and driving path of the driverless vehicle, predict the state of the driverless vehicle in the future for a period of time. The prediction process can consider the dynamic characteristics of the driverless vehicle and external interference factors.

[0115] Furthermore, the control input can be optimized. For example, a cost function can be designed based on the predicted future state and driving path. This cost function usually takes into account factors such as the deviation of the autonomous vehicle from the driving path and the change of the control input. The minimum value of the cost function is obtained by using an optimization algorithm (such as quadratic programming), and the optimal control input (such as throttle opening, steering angle, etc.) is obtained.

[0116] During the actual driving process, the above prediction and optimization processes are continuously repeated. According to the real-time state of the autonomous vehicle and the changes in the external environment, the control input is adjusted in real time, so that the autonomous vehicle can accurately follow the driving path and finally reach the target point.

[0117] In the embodiments of the present application, during the process of generating a driving path and controlling the autonomous vehicle to drive according to the driving path, the target object is searched. On the one hand, it can avoid the collision of the autonomous vehicle with obstacles. On the other hand, it can reach the target point in a shorter time, thereby improving the search efficiency of the target object.

[0118] In some alternative implementation manners, refer to Figure 7 , Figure 7 A flowchart of another target object search method is provided, which specifically includes the following steps:

[0119] S701, determine whether the autonomous vehicle has searched for the target object according to the current environmental information of the autonomous vehicle. If so, execute S709; if not, execute S712.

[0120] S702, obtain the current pose information of the autonomous vehicle.

[0121] S703, obtain the image information and laser point cloud information of the autonomous vehicle in the current search area.

[0122] S704, generate a spatial occupancy grid map of the autonomous vehicle in the current search area according to the image information, laser point cloud information of the autonomous vehicle in the current search area, and the current pose information.

[0123] S705, splice the spatial occupancy grid map of the autonomous vehicle in the current search area with the map corresponding to the searched area to obtain the global map of the search area.

[0124] S706, perform grid division on the global map and the expanded area.

[0125] S707, during the process of expanding the area starting from the current position, if it is determined according to the global map and the expanded area that the grid before expansion belongs to the detected area, and the grid after expansion belongs to the undetected area, then the grid before expansion is used as the boundary point.

[0126] S708. For each boundary point, perform a weighted operation on the searched range ratio, heading angle difference, and position spacing corresponding to the boundary point to obtain a cost function value.

[0127] S709. Take the boundary point corresponding to the minimum cost function value among the boundary points as the target point.

[0128] S710. Generate a driving path from the current position to the target point subject to the path generation conditions.

[0129] S711. During the process of controlling the unmanned vehicle to travel along the driving path, search for the target object.

[0130] S712. End the search task.

[0131] For the specific processes of S701 to S712 above, reference can be made to the description of the method embodiments above. Their implementation principles and technical effects are similar, and will not be elaborated here.

[0132] Moreover, the execution order between the above steps is only an exemplary illustration and is not used to limit the execution steps. Other execution orders are within the protection scope of the embodiments of the present application.

[0133] It should be understood that although the steps in the flowcharts involved in the above embodiments are shown in sequence according to the arrows, these steps do not necessarily need to be executed in the order indicated by the arrows. Unless clearly stated in this article, there is no strict order limit for the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily need to be executed at the same time, but can be executed at different times. The execution order of these steps or stages does not necessarily need to be sequential, but can be executed alternately or in turn with at least some of the steps or stages in other steps or other steps.

[0134] Based on the same inventive concept, the embodiments of the present application also provide a target object search device for implementing the above-mentioned target object search method. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the target object search device provided below can refer to the limitations on the target object search method in the above text, and will not be elaborated here.

[0135] In an exemplary embodiment, as Figure 8 shown, a target object search device is provided, including:

[0136] An acquisition module 10, configured to acquire the current pose information of the driverless vehicle when it is determined according to the current environmental information of the driverless vehicle that the target object has not been searched for;

[0137] A first determination module 20, configured to determine the boundary points of the unsearched area according to the current environmental information, the searched area, and the current pose information;

[0138] A second determination module 30, configured to determine the target point according to the searched range ratio corresponding to the boundary point, the heading angle difference, and the position spacing; the searched range ratio is the ratio between the searched area and the preset area within the preset area corresponding to the boundary point; the heading angle difference is the difference between the heading angle of the driverless vehicle at the current position and the target angle, and the target angle is the angle between the direction from the current position to the boundary point and the reference direction; the position spacing is the spatial distance between the current position and the boundary point;

[0139] A search module 40, configured to control the driverless vehicle to search for the target object during the process of driving from the current position to the target point.

[0140] The above target object search device, when it is determined according to the current environmental information of the driverless vehicle that the target object has not been searched for, acquires the current pose information of the driverless vehicle; and determines the boundary points of the unsearched area according to the current environmental information of the driverless vehicle, the searched area, and the current pose information; determines the target point according to the searched range ratio corresponding to the boundary point, the heading angle difference, and the position spacing; and further controls the driverless vehicle to search for the target object during the process of driving from the current position to the target point. In the above solution, the target point is determined according to the searched range ratio corresponding to the boundary point, the heading angle difference, and the position spacing, so that the determined target point is, relative to the driverless vehicle, the boundary point with the smallest cost among the boundary points that the driverless vehicle reaches from the current position. In this way, the time for the driverless vehicle to reach the next target point for searching can be saved, thereby improving the search efficiency of the target object.

[0141] In one embodiment, the second determination module 30 is specifically configured to:

[0142] For each boundary point, perform a weighted operation on the searched range ratio corresponding to the boundary point, the heading angle difference, and the position spacing to obtain a cost function value; and use the boundary point corresponding to the smallest cost function value among the boundary points as the target point.

[0143] In one embodiment, the number of boundary points is at least two; the first determination module 20 specifically includes:

[0144] A building unit, configured to build a global map of the search area according to the current environmental information, the map corresponding to the searched area, and the current pose information; the search area includes the searched area and the current search area;

[0145] A determination unit, configured to determine boundary points of an unsearched area according to a global map and the expanded area during the process of expanding an area starting from the current position.

[0146] In one embodiment, the current environmental information includes image information and laser point cloud information of the current search area, and the establishment unit is specifically configured to:

[0147] Obtain the image information and laser point cloud information of the unmanned vehicle in the current search area; generate a spatial occupancy grid map of the unmanned vehicle in the current search area according to the image information and laser point cloud information of the unmanned vehicle in the current search area and the current pose information; splice the spatial occupancy grid map of the unmanned vehicle in the current search area with the map corresponding to the searched area to obtain the global map of the search area.

[0148] In one embodiment, the determination unit is specifically configured to:

[0149] Perform grid division on the global map and the expanded area; during the process of expanding the area starting from the current position, if it is determined according to the global map and the expanded area that the grid before expansion belongs to the detected area and the grid after expansion belongs to the undetected area, then use the grid before expansion as the boundary point.

[0150] In one embodiment, the search module 40 is specifically configured to:

[0151] Generate a driving path from the current position to the target point subject to the path generation condition; the path generation condition is that the driving path is the shortest in distance and there are no obstacles in the driving path; search for the target object during the process of controlling the unmanned vehicle to drive according to the driving path.

[0152] Each module in the above target object search device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to the above respective modules.

[0153] In an exemplary embodiment, a computer device is provided. The computer device can be an intelligent driving controller in a vehicle, and its internal structure diagram can be as Figure 9As shown. The computer device includes a processor, a memory, an input / output (I / O) interface, and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store environmental data and vehicle data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a method for searching for a target object.

[0154] Those skilled in the art can understand that Figure 9 the structure shown in is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0155] In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, it implements the steps of the method for searching for a target object described in any of the above embodiments.

[0156] In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, it implements the steps of the method for searching for a target object described in any of the above embodiments.

[0157] In an embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, it implements the steps of the method for searching for a target object described in any of the above embodiments.

[0158] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0159] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in this application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., without limitation.

[0160] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope recorded in this application.

[0161] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A target object search method, characterized in that: The method comprises: When it is determined according to the current environment information of the unmanned vehicle that the unmanned vehicle has not searched for the target object, obtaining the current position information of the unmanned vehicle; Determine the boundary points of the unsearched area according to the current environment information, the searched area, and the current posture information; Determine the target point according to the searched range proportion, heading angle difference and position spacing corresponding to the boundary point; the searched range proportion is the ratio between the searched area and the preset area in the preset area corresponding to the boundary point; the heading angle difference is the difference between the heading angle of the current position of the unmanned vehicle in the current posture information and the target angle, and the target angle is the angle between the direction of the current position pointing to the boundary point and the reference direction; the position spacing is the spatial distance between the current position and the boundary point; The unmanned vehicle is controlled to search for the target object during the process of traveling from the current position to the target point.

2. The method according to claim 1, characterized in that The number of the boundary points is at least two; and the target point is determined according to the search range proportion, heading angle difference and position spacing corresponding to the boundary points, including: For each boundary point, weight calculation is performed on the search range proportion, heading angle difference and position spacing corresponding to the boundary point to obtain a cost function value; The boundary point corresponding to the minimum cost function value among the boundary points is taken as the target point.

3. The method according to claim 1, characterized in that The step of determining the boundary points of the unsearched area according to the current environment information, the searched area, and the current posture information includes: Establishing a global map of the search area according to the current environment information, the map corresponding to the searched area, and the current posture information; the search area includes the searched area and the current search area; In the process of expanding the area with the current position as the starting point, the boundary points of the unsearched area are determined according to the global map and the expanded area.

4. The method according to claim 3, characterized in that The current environment information includes image information and laser point cloud information of the current search area, and establishing a global map of the search area according to the current environment information, a map corresponding to the searched area, and the current posture information includes: Obtaining image information and laser point cloud information of the unmanned vehicle in the current search area; Generate a space occupancy grid map of the unmanned vehicle in the current search area according to the image information and laser point cloud information of the unmanned vehicle in the current search area, as well as the current posture information; The grid map of the space occupied by the unmanned vehicle in the current search area is spliced ​​with the map corresponding to the searched area to obtain a global map of the search area.

5. The method according to claim 4, characterized in that In the process of expanding the area with the current position as the starting point, determining the boundary points of the unsearched area according to the global map and the expanded area includes: Dividing the global map and the extended area into grids; In the process of expanding the area starting from the current position, if it is determined according to the global map and the expanded area that the grid before expansion belongs to the detected area and the grid after expansion belongs to the undetected area, the grid before expansion is used as a boundary point.

6. The method according to claim 1, characterized in that The controlling the unmanned vehicle to search for the target object during the process of traveling from the current position to the target point includes: Generate a driving path from the current position to the target point with a path generation condition as a constraint; the path generation condition is that the driving path has the shortest distance and there are no obstacles in the driving path; The target object is searched for while the unmanned vehicle is controlled to travel along the travel path.

7. A target object search device, characterized in that: The device comprises: An acquisition module, used for acquiring current position information of the unmanned vehicle when it is determined that the unmanned vehicle has not searched for the target object according to the current environment information of the unmanned vehicle; A first determination module is used to determine the boundary points of the unsearched area according to the current environment information, the searched area, and the current posture information; The second determination module is used to determine the target point according to the searched range proportion, heading angle difference and position spacing corresponding to the boundary point; the searched range proportion is the ratio between the searched area and the preset area in the preset area corresponding to the boundary point; the heading angle difference is the difference between the heading angle of the unmanned vehicle at the current position and the target angle, and the target angle is the angle between the direction of the current position pointing to the boundary point and the reference direction; the position spacing is the spatial distance between the current position and the boundary point; The search module is used to control the unmanned vehicle to search for the target object during the process of traveling from the current position to the target point.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images