A method for updating a path planning template library, a terminal device, and a storage medium

Abstract

The application provides an updating method of a planning path template library, a terminal device and a storage medium. The method comprises the following steps: performing region division and path planning according to a preset task region and an initial position set, so as to obtain a cluster planning path corresponding to a cluster of unmanned devices; wherein the initial position set is an initial position corresponding to each unmanned device in the cluster of unmanned devices; the cluster planning path comprises a target planning path corresponding to each unmanned device; a cluster planning parameter corresponding to the cluster planning path is determined according to position information of a target patrol key point in each target planning path; and when it is determined according to the cluster planning parameter that there is no planning path template identical to the cluster planning path in a planning path template library, the cluster planning path is added to the planning path template library. The method can improve the efficiency of updating the planning path template library on the basis of reducing the matching redundancy of the planning path template library.

Description

Technical Field

[0001] This application relates to the field of path planning technology, and in particular to a method for updating a path planning template library, a terminal device, and a storage medium. Background Technology

[0002] With the gradual maturation of autonomous driving technology, utilizing swarms of autonomous vehicles to collaboratively search or measure pre-defined task areas with known environmental parameters has become an increasingly popular research topic in the field. Regarding path planning for swarms of autonomous vehicles executing corresponding tasks within a pre-defined task area, current technical solutions involve pre-establishing a path planning template library. In practical applications, a matching path planning template is determined from this library based on the initial positions of each autonomous vehicle in the swarm. The swarm then performs a coverage search and executes the corresponding task according to the target path corresponding to each autonomous vehicle within that template.

[0003] During the updating of the planning path template library, considering redundancy and matching efficiency, when determining whether to add a determined cluster planning path as a new planning path template to the library, it is necessary to avoid adding cluster planning paths similar to existing planning path templates in the library. At the algorithm execution level, it is difficult to efficiently compare whether a cluster planning path is identical to a planning path template. Traditional technical solutions determine whether to add a cluster planning path to the planning path template library by comparing the similarity of factors such as the initial position, destination position, and trajectory shape of each autonomous vehicle in the cluster. However, this technical solution involves complex calculations, resulting in low efficiency in updating the planning path template library.

[0004] Therefore, how to improve the efficiency of updating the planning path template library is a technical problem that needs to be solved by those skilled in the art. Summary of the Invention

[0005] The purpose of this application is to provide a method and apparatus for updating a planning path template library, a method and apparatus for cluster planning path matching of an unmanned driving equipment cluster, a method and apparatus for coverage search of an unmanned driving equipment cluster, a terminal device, and a computer-readable storage medium, with the aim of improving the efficiency of updating the planning path template library.

[0006] Firstly, this application provides a method for updating a planned path template library. The method includes:

[0007] Based on the preset task area and initial position set, the region is divided and path planning is performed to obtain the cluster planning path corresponding to the autonomous driving equipment cluster; wherein, the initial position set is the initial position corresponding to each autonomous driving equipment in the autonomous driving equipment cluster; the cluster planning path includes the target planning path corresponding to each of the autonomous driving equipment.

[0008] Determine the cluster planning parameters corresponding to the cluster planning path based on the location information of the key points of the target inspection in each target planning path;

[0009] If, based on the cluster planning parameters, it is determined that there is no planning path template in the planning path template library that is identical to the cluster planning path, the cluster planning path is added to the planning path template library.

[0010] This application provides a method for updating a planning path template library. It uses cluster planning parameters calculated based on the location information of key points in each target planning path to determine whether there is a planning path template in the planning path template library that is the same as the cluster planning path. This process is more efficient and convenient than the existing method of judging by calculating the similarity of factors such as the initial position, destination position, and trajectory shape of each unmanned vehicle. Therefore, this method can improve the efficiency of updating the planning path template library while reducing the matching redundancy of the planning path template library.

[0011] It is understood that the path planning template library update device, the cluster path planning matching method and device for an unmanned driving equipment cluster, the coverage search method and device for an unmanned driving equipment cluster, the terminal device and the computer-readable storage medium provided in the embodiments of this application have the same beneficial effects as the path planning template library update method described above, and will not be repeated here. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0013] Figure 1 A flowchart illustrating a method for updating a path planning template library as provided in an embodiment of this application;

[0014] Figure 2In the update method of the planning path template library provided in the embodiment of this application, S100: the region is divided and the path is planned according to the preset task area and the initial position set to obtain a flowchart of the cluster planning path corresponding to the unmanned driving equipment cluster;

[0015] Figure 3 S130 of another path planning method for an unmanned vehicle cluster provided in this application embodiment: Iteratively constructing search area combinations based on an initial location set, and determining the target search area combination based on the area division score of each search area combination;

[0016] Figure 4 A schematic diagram of an interior / exterior angle and a maximum distance provided for an embodiment of this application;

[0017] Figure 5 S132 of the path planning method for another unmanned vehicle cluster provided in this application embodiment: a schematic diagram of the process of adjusting the region based on the initial search region combination to obtain the candidate search region combination;

[0018] Figure 6 S140 of another path planning method for an unmanned vehicle cluster provided in this application embodiment: For each unmanned vehicle, a planned path is iteratively constructed based on the target search area, and the target planned path of the unmanned vehicle is determined according to the path planning score of each planned path.

[0019] Figure 7 A schematic diagram of an initial planned path provided for an embodiment of this application;

[0020] Figure 8 A schematic diagram illustrating a target planning path corresponding to each unmanned driving device, provided for embodiments of this application;

[0021] Figure 9 A flowchart for determining the path planning score corresponding to each planned path is provided in this application embodiment;

[0022] Figure 10 A schematic diagram illustrating the process of calculating a planned path score, provided in an embodiment of this application;

[0023] Figure 11 A schematic diagram illustrating the relationship between a preset task area, a task area grid, a sensing radius, a coverage radius, and key patrol points is provided for an embodiment of this application.

[0024] Figure 12 This is a schematic diagram illustrating how each unmanned driving device travels along a corresponding target planned path within a preset task area, as provided in an embodiment of this application.

[0025] Figure 13A schematic diagram illustrating the location information of key inspection points provided in this application embodiment;

[0026] Figure 14 This is a schematic diagram illustrating a process for determining cluster planning parameters corresponding to a cluster planning path based on the location information of key points in each target planning path, as provided in an embodiment of this application.

[0027] Figure 15 This is a schematic diagram illustrating the structure of a cluster planning path storage format in a planning path template library provided in an embodiment of this application.

[0028] Figure 16 A flowchart of a cluster planning path matching method for an unmanned vehicle cluster provided in this application embodiment;

[0029] Figure 17 A flowchart of a cluster planning path matching method for an unmanned vehicle cluster provided in this application embodiment;

[0030] Figure 18 This is a schematic diagram illustrating the process of determining the matching cluster planning path corresponding to the unmanned driving equipment cluster to be matched in the planning path template library based on the number of equipment and the set of key points to be matched for inspection, as provided in this application embodiment.

[0031] Figure 19 A flowchart illustrating a coverage search method for an unmanned vehicle cluster provided in this application embodiment. Detailed Implementation

[0032] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.

[0033] This application provides a method for updating a path planning template library, which can be executed by the processor of a terminal device when running a corresponding computer program.

[0034] Figure 1 The flowchart illustrates a method for updating a path planning template library as provided in this application embodiment. For ease of explanation, only the parts relevant to this embodiment are shown. The method provided in this embodiment includes the following steps:

[0035] S100: Based on the preset task area and initial position set, perform area division and path planning to obtain the cluster planning path corresponding to the autonomous driving equipment cluster; wherein, the initial position set is the initial position corresponding to each autonomous driving equipment in the autonomous driving equipment cluster; the cluster planning path includes the target planning path corresponding to each autonomous driving equipment.

[0036] It should be noted that the initial position set can be determined by randomly determining the initial position of each autonomous vehicle in the autonomous vehicle cluster. By repeatedly determining the initial position set and generating the corresponding cluster planning path, the planning path template library is continuously updated.

[0037] S200: Determine the cluster planning parameters corresponding to the cluster planning path based on the location information of key points of target inspection in each target planning path.

[0038] Among them, the cluster planning parameters refer to the parameters that characterize the features of the cluster planning path. In this embodiment, the cluster planning parameters are determined based on the location information of the key points of the target inspection in each target planning path in the cluster planning path.

[0039] Specifically, after determining the cluster planning path, the target inspection key points corresponding to each unmanned vehicle in the cluster planning path are obtained, and the location information of each target inspection key point is obtained. Then, the cluster planning parameters corresponding to the cluster planning path are calculated based on the location information. The location information can be the coordinate position of the target inspection key point or the latitude and longitude position of the target inspection key point. This embodiment does not limit this.

[0040] S300: If, based on the cluster planning parameters, there is no planning path identical to the cluster planning path in the planning path template library, the cluster planning path will be added to the planning path template library.

[0041] Specifically, after determining the cluster planning parameters, it checks if a cluster planning parameter identical to the given cluster planning parameter exists in the planning template library. If the same cluster planning parameter exists in the planning template library, it means that a planning path template with the same cluster planning path exists in the planning template library, so the cluster planning path is not added again to the planning template library. If the same cluster planning parameter does not exist in the planning template library, it means that a cluster planning path corresponding to the given cluster planning parameter does not exist in the planning template library, that is, a planning path template with the same cluster planning path does not exist in the planning path template library, so the cluster planning path is added as a new planning path template to the planning path template library, and the planning path template library is updated. Specifically, the cluster planning path includes the target planning path corresponding to each autonomous driving device. The target planning path includes the initial position, target inspection key points, and traversal order of the target inspection key points for each autonomous driving device.

[0042] This application provides a method for updating a planning path template library. It uses cluster planning parameters calculated based on the location information of key points in each target planning path to determine whether there is a planning path template in the planning path template library that is the same as the cluster planning path. This process is more efficient and convenient than the existing method of calculating the similarity of factors such as the initial position, destination position, and trajectory shape of each unmanned vehicle. Therefore, this method can improve the efficiency of updating the planning path template library while reducing the matching redundancy of the planning path template library.

[0043] Figure 2 In the method for updating a planning path template library provided in this application embodiment, S100 involves: performing region division and path planning based on a preset task area and an initial location set to obtain a flowchart of a cluster planning path corresponding to an unmanned vehicle cluster; as shown in Figure 2, the process of performing region division and path planning based on a preset task area and an initial location set to obtain a cluster planning path corresponding to an unmanned vehicle cluster includes:

[0044] S110: Determine the set of key inspection points corresponding to the preset task area of ​​the unmanned vehicle cluster, and the inspection coverage area corresponding to each key inspection point in the set of key inspection points.

[0045] The unmanned equipment can be unmanned surface vessels (USVs), drones, or unmanned vehicles, etc., and this embodiment does not limit the specific type of unmanned equipment. The preset task area refers to the area that the USV cluster needs to cover during task execution, and the task types include searching, surveying, and mapping. In practice, determining the preset task area includes setting the shape, length, width, and the location, shape, and area of ​​obstacles within the preset task area according to the actual task scenario. The key points for inspection refer to the locations corresponding to the preset task area that the USVs cover to the maximum extent, and are generally set in locations that the USVs can reach. The USV's search, surveying, or data collection tasks within the preset water area can be equivalently converted into a process of traversing the key points within the preset water area.

[0046] In practice, the key points for inspection are first determined based on the preset task area. Then, the coverage radius corresponding to the key points for inspection is determined based on the perception radius of the unmanned driving equipment. Finally, the square area centered on the key points for inspection and with the coverage radius as the radius is determined as the inspection coverage area.

[0047] S120: Obtain the initial position set corresponding to the cluster of unmanned vehicles; wherein, the initial position set includes the initial position corresponding to each unmanned vehicle.

[0048] The initial position refers to the location of each autonomous vehicle in the autonomous vehicle cluster within the pre-defined task area. In actual operation, the initial positions of each autonomous vehicle are distributed as close as possible to the pre-defined task area to efficiently achieve a comprehensive search of the area.

[0049] In actual operation, the unmanned driving equipment cluster is initialized, the number of each unmanned driving equipment in the cluster is determined, the driving speed and perception radius of each unmanned driving equipment are set, and the initial position of each unmanned driving equipment is randomly configured in the preset task area. The area corresponding to the initial position is a drivable area.

[0050] S130: Iteratively construct search area combinations based on the initial location set, and determine target search area combinations based on the area division scores of each search area combination; wherein, the target search area combination includes target search areas corresponding to each unmanned driving device, and the target search area includes at least one patrol coverage area.

[0051] It should be noted that when a large number of autonomous vehicles participate in the task, a corresponding search area needs to be allocated to each autonomous vehicle to ensure that each vehicle can perform the task efficiently. Specifically, after determining the initial positions of each autonomous vehicle in the cluster, an initial search area combination is determined based on each initial position. Candidate search area combinations are then iteratively constructed based on these initial search area combinations, and a region division score is determined for each search area combination. The search area combination includes the initial search area combination and each candidate search area combination. Each search area combination includes search areas corresponding to each autonomous vehicle, and each search area includes at least one patrol coverage area. Then, a target search area combination is determined based on the region division score, and each search area in this target search area combination is the target search area corresponding to each autonomous vehicle.

[0052] S140: For each autonomous vehicle, iteratively construct a planned path based on the target search area, and determine the target planned path of the autonomous vehicle based on the path planning score of each planned path.

[0053] Specifically, after determining the target search area corresponding to each autonomous driving device, for each autonomous driving device, the initial planned path is obtained by connecting the key inspection points corresponding to the target search area in the order of their completion and termination. Candidate planned paths are then constructed based on the initial planned path. The candidate planned paths are then used as updated initial planned paths, and candidate planned paths are constructed again based on the updated initial planned paths. The path planning score corresponding to each planned path is determined. The planned path includes the initial planned path and each candidate planned path. Finally, the target planned path corresponding to the autonomous driving device is determined based on the path planning score.

[0054] After performing the above operations on each autonomous driving device, the target planning path corresponding to each autonomous driving device is obtained.

[0055] In this embodiment, by performing dual iterative optimization on the combination of search areas of the autonomous driving equipment cluster and the planned path of the autonomous driving equipment, the target planned path corresponding to each autonomous driving equipment is determined. Therefore, when each autonomous driving equipment in the autonomous driving cluster performs its corresponding task according to the corresponding target planned path, the work efficiency of the autonomous driving equipment in performing the task can be improved.

[0056] Figure 3In another path planning method for an unmanned vehicle cluster provided in this application embodiment, S130 is a flowchart of iteratively constructing search area combinations based on an initial position set and determining a target search area combination based on the area division score of each search area combination. Based on the above embodiment, this embodiment further explains and optimizes the technical solution. Specifically, in this embodiment, S130: iteratively constructing search area combinations based on an initial position set and determining a target search area combination based on the area division score of each search area combination includes:

[0057] S131: Determine the initial search area combination based on the initial location set.

[0058] The initial search area is determined directly based on the initial position of the autonomous vehicle, and includes at least one patrol coverage area. The combination of initial search areas includes initial search areas corresponding to each autonomous vehicle. Specifically, the distances between the autonomous vehicle and each patrol key point are calculated based on its initial position. Following a proximity allocation principle, a preset number of patrol key points closest to the autonomous vehicle are determined, and the patrol coverage area corresponding to these determined patrol key points is defined as the initial search area for that autonomous vehicle.

[0059] S132: Adjust the region based on the initial search region combination to obtain the candidate search region combination.

[0060] Specifically, the area adjustment operation includes modifying, adding, or deleting the patrol coverage area in the initial search area combination, and a new candidate search area combination is obtained after each adjustment.

[0061] In one specific embodiment, a pre-set upper limit for the region partitioning population size corresponding to the autonomous driving device cluster is set; the upper limit for the region partitioning population size refers to the maximum number of search region combinations determined during a single iteration of construction. Generally, the value of the upper limit for the region partitioning population size is positively correlated with the number of autonomous driving devices in the autonomous driving device cluster.

[0062] It should be noted that after obtaining each candidate search region combination, the candidate search region combination is added to the region partitioning population, and the size of the region partitioning population is updated, i.e., the region partitioning population is expanded; it is then determined whether the size of the region partitioning population has reached the upper limit of the region partitioning population size; if it has not reached the upper limit of the region partitioning population size, the region is adjusted based on the initial search region combination to obtain candidate search region combinations, and the region partitioning population is expanded; if the upper limit of the region partitioning population size is reached, the expansion of the region partitioning population ends and the region partitioning population is output.

[0063] S133: Determine the region division score corresponding to each search region combination.

[0064] Among them, the regional division score refers to the score used to characterize the quality of the combination of search regions.

[0065] S134: Filter the combinations of search areas based on the scores of each area division.

[0066] For example, if the region segmentation score is positively correlated with the overall integrity of the search area, when filtering search area combinations based on each region segmentation score, the search area combinations with higher region segmentation scores are selected, meaning that search area combinations with lower region segmentation scores are deleted first. Specifically, search area combinations are deleted according to a preset ratio or a specified number, and search area combinations that meet the region segmentation score requirements are selected.

[0067] S135: Determine whether the completion condition of the first iteration has been met;

[0068] S136: If the conditions for ending the first iteration have not been met, update the selected search region combinations to the initial search region combinations and return to S132: the step of adjusting the regions based on the initial search region combinations to obtain candidate search region combinations.

[0069] S137: If the first iteration construction end condition is met, the target search area corresponding to each autonomous driving device is determined, and a target search area combination is obtained. The first iteration construction end condition can be that the number of iterations reaches the first iteration limit. The value of the first iteration limit is positively correlated with the number of autonomous driving devices in the autonomous driving device cluster. In this embodiment, the process of constructing a search area combination in one iteration is the process of "region division population expansion → region division population scoring → region division population reduction"; for example, the entire process of expanding the region division population from 100 to 1000 and then reducing it back to 100 is called the process of constructing a search area combination in one iteration.

[0070] Specifically, the process involves determining whether the first iteration construction end condition has been met. If not, the selected search region combinations are updated to the initial search region combinations, and the process returns to S300: This involves adjusting the regions based on the initial search region combinations to obtain candidate search region combinations, i.e., continuing the iterative construction operation based on the current region partitioning population. If the first iteration construction end condition has been met, the current region partitioning population is determined, and then the search region combination with the optimal region partitioning score is determined from the current region partitioning population according to the scoring rules; this is the determination of the target search region combination. The target search region combination includes the target search regions corresponding to each autonomous driving device; therefore, the target search regions corresponding to each autonomous driving device are determined based on the target search region combination.

[0071] According to the method in this embodiment, by iteratively constructing combinations of search areas, the optimal target search area corresponding to each autonomous driving device can be determined.

[0072] Based on the above embodiments, this embodiment further explains and optimizes the technical solution. Specifically, in this embodiment, S132: determining the region division score corresponding to each search region combination, including:

[0073] For each search area combination, determine the search area corresponding to each autonomous driving device; determine the total number of interior and exterior angles of the search area combination based on each search area, and obtain the interior and exterior angle score; determine the maximum distance corresponding to each search area, and determine the average value of each maximum distance, and obtain the interval score; determine the area division score corresponding to the search area combination based on the interior and exterior angle score and the interval score.

[0074] Figure 4 This application provides a schematic diagram of interior and exterior angles and maximum distance, as part of an embodiment. The interior and exterior angles refer to angles of 90° or 270° within the search area. Figure 4 In the search area corresponding to the autonomous driving equipment, the angles circled by circles are interior and exterior angles, while the angles not circled are non-interior and exterior angles. For each combination of search areas, firstly, the search areas corresponding to each autonomous driving equipment in the combination are determined; then, the number of interior and exterior angles corresponding to each search area is determined, and the total number of interior and exterior angles corresponding to the combination of search areas is calculated, thus determining the interior and exterior angle score.

[0075] In one specific embodiment, the total number of interior and exterior angles of the search area combination is determined based on each search area to obtain an interior and exterior angle score, including:

[0076] For each search area, determine the coordinates corresponding to the four corners of each patrol coverage area, and determine the number of repetitions corresponding to each coordinate. If the number of repetitions is odd, the angles corresponding to the number of repetitions are determined as interior and exterior angles, and the number of interior and exterior angles of the search area is determined. Based on the number of interior and exterior angles corresponding to each search area, determine the total number of interior and exterior angles corresponding to the combination of search areas, and obtain the interior and exterior angle score.

[0077] For example, regarding the search area corresponding to unmanned vehicle 1 (USV1), the coordinate of the upper left corner of the patrol coverage area 1 is repeated 1 time, which is an odd number, therefore this corner corresponds to an interior / exterior angle; the coordinate of the upper right corner of the patrol coverage area 1 overlaps with the coordinate of the upper left corner of the patrol coverage area 2, and this coordinate is repeated 2 times, which is an even number, therefore this corner is not an interior / exterior angle; the coordinates of the lower right corner of the patrol coverage area 1, the lower left corner of the patrol coverage area 2, and the upper left corner of the patrol coverage area 10 overlap, and this coordinate is repeated 3 times, which is an odd number, therefore this corner corresponds to an interior / exterior angle. Following the above method, each corner in the search area is sequentially determined to be an interior / exterior angle, and the number of interior / exterior angles in the search area is counted; then, the number of interior / exterior angles corresponding to each search area in the search area combination is accumulated to calculate the total number of interior / exterior angles, thus obtaining the interior / exterior angle score.

[0078] Specifically, for each search area, the distance between any two patrol coverage areas within the search area is determined, and the maximum distance is determined based on each distance, i.e., the maximum distance corresponding to the search area is determined; then, for each combination of search areas, the average of the maximum distances of each search area in the combination is calculated to obtain the interval score.

[0079] After determining the interior and exterior angle scores and interval scores corresponding to the search area combination, the region division score corresponding to the search area combination is calculated based on the interior and exterior angle scores and interval scores; specifically, the region division score = interior and exterior angle score × interval score. In a specific embodiment, it is assumed that the number of autonomous vehicles in the autonomous vehicle cluster is n, and the number of interior and exterior angles of the search area corresponding to the i-th autonomous vehicle is s. i The maximum distance of the search area corresponding to the i-th unmanned vehicle is d. i Then the region division score f corresponding to the search region combination A for:

[0080]

[0081] According to the method of this embodiment, a region division score is determined for each search region combination, and the region division score can accurately describe the integrity of the search region combination.

[0082] In one specific embodiment, region adjustment is performed based on the initial search region combination to obtain a candidate search region combination, including: swapping the patrol coverage areas in different initial search regions in the initial search region combination to obtain a candidate search region combination.

[0083] Specifically, after determining the initial search area combination, any two autonomous driving devices are selected, and at least one patrol key point is selected from the patrol key point set corresponding to the two autonomous driving devices. The selected patrol key points are swapped, which changes the search area of ​​the autonomous driving devices. Therefore, a search area combination different from the initial search area combination is obtained, which is the candidate search area combination.

[0084] Figure 5 S132 of the path planning method for an unmanned vehicle cluster provided in this application embodiment is a schematic diagram of the process of adjusting the search area based on the initial search area combination to obtain a candidate search area combination. Specifically, in this embodiment, the patrol key point 26 in unmanned vehicle 2 (USV2) and the patrol key point 21 in unmanned vehicle 3 (USV3) are swapped, so that the search areas corresponding to unmanned vehicle 2 (USV2) and unmanned vehicle 3 (USV3) change, and a new search area combination is obtained, that is, a candidate search area combination is obtained.

[0085] In this embodiment, by exchanging the patrol coverage areas in different initial search areas in the initial search area combination, a candidate search area combination is obtained. Compared with the method of directly adding or deleting patrol key points to obtain the candidate search area combination, this method can avoid missing or repeating patrol key points in the candidate search area combination, thus further improving the working efficiency of the unmanned equipment cluster.

[0086] Figure 6 In another path planning method for an unmanned vehicle cluster provided in this application embodiment, S140 is a flowchart illustrating how, for each unmanned vehicle, a planned path is iteratively constructed based on the target search area, and the target planned path for the unmanned vehicle is determined according to the path planning score of each planned path. Based on the above embodiment, this embodiment further explains and optimizes the technical solution. Specifically, in this embodiment, S140, for each unmanned vehicle, iteratively constructs a planned path based on the target search area, and determines the target planned path for the unmanned vehicle according to the path planning score of each planned path, including:

[0087] S141: For each autonomous vehicle, determine the initial planned path based on the target search area of ​​the autonomous vehicle.

[0088] The initial planned path refers to the planned path directly determined based on the target search area of ​​the autonomous driving equipment. In actual operation, the initial planned path can be determined according to the arrangement order of the patrol key points in the target search area; the initial planned path is the traversal order of the autonomous driving equipment through the patrol key points in the target search area in a certain order.

[0089] like Figure 7 The diagram illustrates an initial planning path provided in an embodiment of this application. Each autonomous driving device independently generates its own initial planning path. Key point 1 is the head key point in the initial planning path, and key point 19 is the tail key point. The autonomous driving device travels from its initial position to the head key point and sequentially traverses each key point. After reaching the tail key point, it returns to the head key point to restart the traversal.

[0090] S142: Adjust the path based on the initial planned path to obtain candidate planned paths.

[0091] Specifically, the positions of the two inspection key points in the initial planned path, excluding the inspection key point at the column head, can be randomly swapped, and a new candidate planned path is obtained after each swap.

[0092] In one specific embodiment, a pre-set upper limit for the path planning population size corresponding to each target search area is set. The upper limit refers to the maximum number of planned paths determined during one iteration of construction. Generally, the value of the upper limit for the path planning population size is positively correlated with the number of key points to be inspected in the target search area. It should be noted that after each candidate planned path is obtained, the candidate planned path is added to the path planning population, and the size of the path planning population is updated, i.e., the path planning population is expanded; it is then determined whether the size of the path planning population has reached the upper limit; if it has not reached the upper limit, the path is adjusted based on the initial planned path to obtain candidate planned paths and expand the path planning population; if the upper limit is reached, the expansion of the path planning population ends and the path planning population is output.

[0093] S143: Determine the path planning score corresponding to each planned path.

[0094] The route planning score refers to the numerical value used to characterize the quality of a planned route. Generally, three factors influence the quality of route planning: the length of the planned route, whether the route crosses obstacles, and the number of turns (turns reduce patrol coverage). In practice, if the route planning score is positively correlated with the quality of the route, then the shorter the route, the fewer obstacles it crosses, and the fewer the turns, the higher the route planning score.

[0095] S144: Filter the planned paths based on the planning scores of each path.

[0096] For example, if the route planning score is positively correlated with the quality of the route planning, when selecting planned routes based on each route planning score, the planned routes with higher scores will be selected, that is, the planned routes with lower scores will be deleted first.

[0097] Specifically, the corresponding planned paths are deleted according to a preset ratio or a specified number, and the planned paths that meet the requirements of the path planning score are selected.

[0098] S145: Determine whether the current condition for the end of the second iteration has been met;

[0099] S146: If the conditions for ending the second iteration have not been met, update each selected planning path to the initial planning path and return to S142: the step of adjusting the path based on the initial planning path to obtain the candidate planning path;

[0100] S147: If the second iteration construction end condition is met, determine the target planning path corresponding to the autonomous driving equipment.

[0101] The second iteration can end when the number of iterations reaches the upper limit of the second iteration count. The upper limit of the second iteration count is positively correlated with the number of key points in the target search area.

[0102] In this embodiment, the process of constructing a planned path in one iteration is the process of "expanding the path planning population → scoring the path planning population → reducing the path planning population".

[0103] Specifically, the process involves determining whether the conditions for ending the second iteration have been met. If not, the selected path plans are updated to the initial path plans, and the process returns to S300: the step of adjusting the search area based on the initial search area combination to obtain candidate search area combinations, i.e., continuing the iterative construction operation based on the current path plan population. If the conditions for ending the second iteration have been met, the current path plan population is determined, and then the optimal path plan with the best score is selected from the current path plan population according to the scoring rules, i.e., the target path plan corresponding to the autonomous driving device is determined.

[0104] Figure 8 This application provides a schematic diagram of a target planning path corresponding to each autonomous driving device, as shown in the embodiments of the present application. Figure 8 As shown, each unmanned vehicle travels from its initial position to the inspection key point corresponding to the head of the column in the planned path, and then traverses the corresponding inspection key points in sequence according to the planned path. After traversing to the inspection key point at the tail of the column, it returns to the inspection key point at the head of the column to start the traversal again.

[0105] It should be noted that in actual operation, the autonomous driving equipment is identified in sequence, and the above steps are performed for each autonomous driving equipment to determine the target planning path corresponding to each autonomous driving equipment.

[0106] According to the method in this embodiment, by iteratively constructing the planning path, the optimal target planning path corresponding to each autonomous driving device can be determined.

[0107] Based on the above embodiments, this embodiment further explains and optimizes the technical solution. Specifically, in this embodiment, determining the path planning score corresponding to each planned path includes:

[0108] For each planned path, the initial distance is determined based on the initial position and the inspection key points corresponding to the column headers in the planned path; the driving path length and number of turns corresponding to two adjacent inspection key points in the planned path are determined; and the planned path score is determined based on the initial distance, driving path length, and number of turns.

[0109] Figure 9 This application provides a flowchart for determining the path planning score corresponding to each planned path. Specifically, different planned paths and their corresponding initial positions are predetermined; each planned path includes at least multiple inspection key points; for each planned path, the following operations are performed: determining the initial distance based on the initial position and the inspection key points corresponding to the column headers in the planned path; traversing each inspection key point in the order of the inspection key points corresponding to the planned path to determine the driving path length and number of turns corresponding to two adjacent inspection key points in the planned path; if the traversal is not yet complete, continuing the traversal in the order of the inspection key points corresponding to the planned path; if the traversal is complete, calculating the planned path score based on the initial distance, driving path length, and number of turns.

[0110] Combination Figure 10 The diagram illustrates a process for calculating a planned path score. Specifically, the process of determining the driving path length and number of turns corresponding to two adjacent inspection key points in the planned path includes: determining the inspection key point corresponding to the column head in the planned path; determining the initial position of the unmanned driving device corresponding to the planned path; and calculating the initial distance d0 based on the initial position and the inspection key point corresponding to the column head.

[0111] Specifically, the process of determining the driving path length corresponding to two adjacent key inspection points in the planned route is as follows:

[0112] Calculate the distance between two adjacent inspection key points; determine if there are obstacles between these two points; if obstacles exist, determine the travel path length as 10 times the distance; if no obstacles exist, directly determine the travel path length as the distance itself. The travel path length between the (j-1)th and jth inspection key points is d. j If the number of key points to be inspected in the search path corresponding to the unmanned vehicle is m, then the number of driving path lengths corresponding to the search path is m-1.

[0113] Specifically, the process of determining the number of turns corresponding to two adjacent key inspection points in the planned path is as follows:

[0114] Calculate the closing angle between two adjacent inspection key points; for example, the closing angle θ1 between inspection key point 3 and inspection key point 11 is 0; the closing angle θ2 between inspection key point 11 and inspection key point 19 is 0; and the closing angle θ3 between inspection key point 19 and inspection key point 18 is 270°. Determine if two adjacent sequential connection angles are the same; if they are the same, the planned path is not a turn at the inspection key point; if they are different, the planned path is a turn at the inspection key point, and the turn count is incremented by 1; for example, if the closing angles θ2 and θ3 are different, the position corresponding to inspection key point 19 is a turn, and the turn count is incremented by 1; after traversing the planned path, determine the turn count w corresponding to the planned path. j .

[0115] Specifically, after determining the initial distance, driving path length, and number of turns corresponding to the planned path, the planned path score is calculated based on the initial distance, driving path length, and number of turns.

[0116] In one specific embodiment, the path planning score f is calculated. B as follows:

[0117]

[0118] Where m represents the number of key points to be inspected in the search path; d0 represents the initial distance, and w j d represents the number of turns. j This represents the length of the travel path between the (j-1)th inspection key point and the jth inspection key point.

[0119] According to the method in this embodiment, the path planning score corresponding to each planned path can be accurately calculated.

[0120] In one specific embodiment, combined with Figure 11The diagram shown in this application embodiment illustrates the relationship between a preset task area, a task area grid, a sensing radius, a coverage radius, and key inspection points; as shown... Figure 11 As shown, the black dashed box represents the task area grid, and the black solid box represents the patrol coverage area; the perception radius of the autonomous vehicle is R, and the coverage radius of the patrol coverage area is r. Specifically, S100: The process of determining the set of patrol key points corresponding to the preset task area of ​​the autonomous vehicle cluster and the patrol coverage area corresponding to each patrol key point in the set of patrol key points includes:

[0121] Step 1: Determine the task area grid corresponding to the preset task area of ​​the unmanned vehicle cluster.

[0122] The grid edge length of the task area grid is determined by the accuracy of the acquisition device used to collect the regional state of the preset task area. In order to facilitate the description of the regional state of the preset task area, the grid information corresponding to the task area grid is set to include: the regional grid coordinates, regional properties, search time, and number of repeated searches (initialized to 0). Regional properties include drivable areas and obstacles, such as water or obstacles.

[0123] Step 2: Determine the key point spacing and mark the key points of the patrol based on the preset task area.

[0124] The keypoint spacing of the patrol key points is the side length of the patrol coverage area, which can be determined based on the side length of the task area grid, obstacle positions, obstacle shapes, and obstacle areas. In this embodiment, the keypoint spacing is twice the coverage radius r, meaning the side length of the patrol coverage area is twice the coverage radius r. The term "marking patrol key points" refers to the anchor points within each patrol key point.

[0125] Step 3: Determine the set of patrol key points based on the key point spacing and the calibrated patrol key points, and determine the patrol coverage area corresponding to each patrol key point in the patrol key point set.

[0126] In practice, starting with the calibrated key inspection points, the coordinates of other key inspection points are calculated sequentially within the preset task area based on the key point spacing of the calibrated key inspection points. In other words, the coordinates of other key inspection points are calculated sequentially based on the calibrated key inspection points according to the key point spacing.

[0127] It should be noted that the principle for selecting and calibrating key points for inspection is to ensure that the maximum number of key points can be obtained and effectively traversed; the determined spacing between key points is such that the traversal of key points by each unmanned vehicle can basically cover the preset task area.

[0128] Specifically, taking each key patrol point as the center, and using twice the coverage radius as the side length of the patrol coverage area, the preset task area is divided into horizontal and vertical grids; the grids corresponding to the center points of the grids are retained as drivable areas, thus determining the patrol coverage areas corresponding to each key patrol point.

[0129] Figure 12 This is a schematic diagram of each unmanned driving device traveling in a preset task area according to the corresponding target planning path, provided by an embodiment of this application; as shown in 12, each coordinate is the coordinate of the corresponding inspection key point, and the inspection coverage area with the same shade represents the target search area corresponding to the same unmanned driving device; the arrow indicates the target planning path of the unmanned driving device in the target search area.

[0130] As can be seen, by setting the patrol key points and patrol coverage area according to the method of this embodiment, the patrol coverage area can cover the preset task area to the maximum extent when the unmanned vehicle traverses each patrol key point, thereby realizing the patrol of the preset task area.

[0131] Based on the above embodiments, this embodiment further explains and optimizes the technical solution. Specifically, in this embodiment, the cluster planning parameters corresponding to the cluster planning path are determined according to the location information of the key points of the target inspection in each target planning path, including:

[0132] Determine the location information of the key points of the target inspection for each target planning path in the cluster planning path; for each target planning path, determine the coordinate calculation value corresponding to the target planning path; determine the cluster planning parameters corresponding to the cluster planning path based on the coordinate calculation value corresponding to each target planning path; among which, the cluster planning parameters include the average product value and the average sum value.

[0133] Specifically, first, the location information of the key points of the target inspection in each target planning path in the cluster planning path is determined; then, for each target planning path, the coordinate calculation value corresponding to the target planning path is determined; after determining the coordinate calculation value corresponding to each target planning path, the average product value and average sum value are calculated based on the coordinate calculation value, that is, the cluster planning parameters corresponding to the cluster planning path are obtained.

[0134] More specifically, assume the number of autonomous driving devices is n, meaning the number of target planning paths is n; and the number of target inspection key points corresponding to the i-th autonomous driving device is m. i ; where the center point coordinates of the j-th target inspection key point are (x ij ,y ij If the coordinates P corresponding to the target planned path are calculated, the formula is as follows:

[0135]

[0136] The formula for calculating the average product M of the cluster path planning based on the coordinates of each target planning path is as follows:

[0137]

[0138] The formula for calculating the average sum A of the cluster path planning based on the coordinates of each target planning path is as follows:

[0139]

[0140] In one specific embodiment, reference Figure 13 The diagram shows the location information of key inspection points and Figure 14 This diagram illustrates a process for determining cluster planning parameters corresponding to the cluster planning path based on the location information of target inspection key points in each target planning path. Assuming the autonomous vehicle cluster includes 5 autonomous vehicles, the target inspection key points in the target planning path of each autonomous vehicle are determined, and their coordinates and values ​​are calculated based on the coordinate values ​​of the target inspection key points. For example, for USV1, its corresponding target planning path is: 1→2→3→11→19→18→10. The location information corresponding to each target inspection key point in the target planning path is determined, such as target inspection key point 1 (350, 10), target inspection key point 2 (350, 250), target inspection key point 3 (350, 350)... target inspection key point 10 (250, 250), and the parameters are calculated...

[0141] The corresponding calculated coordinate value P1 = 383.99; the calculated coordinate values ​​for the other unmanned vehicles are calculated using the same method; for example, the calculated coordinate value for USV2 is P2 = 598.17, for USV3 it is P3 = 901.39, for USV4 it is P4 = 254.95, and for USV5 it is P5 = 738.24; the average product value is calculated based on the calculated coordinate values ​​for each unmanned vehicle. Average sum This yields the cluster planning parameters corresponding to the cluster planning path.

[0142] As can be seen, this embodiment uses the average product and average sum of the cluster planning paths as the deduplication index of the planning path template library to avoid storing similar planning path templates in the planning path template library; according to the method of this embodiment, the cluster planning parameters corresponding to the cluster planning path can be calculated efficiently and conveniently, and the efficiency of updating the planning path template library can be improved.

[0143] Based on the above embodiments, this embodiment further explains and optimizes the technical solution. Specifically, in this embodiment, a method for updating a planning path template library further includes: if the number of cluster planning paths in the planning path template library reaches a preset number threshold, and / or the duration of creating the planning path template library reaches a preset duration threshold, the planning path template library is output.

[0144] Specifically, in one embodiment, after each update of the planned path template library, the number of templates corresponding to the pre-set template quantity variable is incremented by 1. Then, it is determined whether the updated template quantity is greater than or equal to a preset quantity threshold. If the template quantity is greater than or equal to the preset quantity threshold, that is, the number of cluster planned paths in the planned path template library has reached the preset quantity threshold, the planned path template library is output. In another embodiment, the time taken to create the planned path template library is obtained. Specifically, the time can be started from when the preset task area is obtained, or from when the cluster parameters of the autonomous driving equipment cluster are obtained, to obtain the time taken to create the planned path template library. It is then determined whether the time taken to create the planned path template library has reached a preset time threshold. If the preset time threshold is reached, it means that the prescribed time limit for establishing the planned path template library has been reached, and therefore the planned path template library is output.

[0145] If the number of cluster planning paths in the planning path template library does not reach the preset number threshold, or the time for creating the planning path template library does not reach the preset time threshold, it means that the planning path template library needs to be updated. Therefore, the corresponding cluster planning path is determined according to different initial location sets, that is, the step of "dividing the region and planning the path according to the preset task area and the initial location set to obtain the cluster planning path corresponding to the unmanned vehicle cluster" is returned.

[0146] In other embodiments, the planned path template library can be output when both the quantity and duration requirements are met. That is, the planned path template library is output when the number of cluster planned paths in the planned path template library reaches a preset quantity threshold and the duration of creating the planned path template library reaches a preset duration threshold.

[0147] In this embodiment, when the number of cluster planning paths in the planning path template library reaches a preset number threshold, and / or the time taken to create the planning path template library reaches a preset time threshold, the planning path template library is output, thus determining a planning path template library that meets the requirements.

[0148] Based on the above embodiments, this embodiment further explains and optimizes the technical solution. Specifically, in this embodiment, a method for updating a planning path template library further includes:

[0149] The cluster planning path corresponding to the unmanned vehicle cluster is stored using vector storage; the location information of each target inspection key point and the unmanned vehicle corresponding to each target inspection key point are stored using map storage.

[0150] Among them, vector (capacity) is a sequence container that encapsulates a dynamically sized array. It can store objects of various types. The objects stored in the sequence container are ordered in a strict linear order, and the corresponding stored object can be accessed by its position in the sequence.

[0151] A map is an associative container that stores keys and values ​​together, allowing users to quickly look up the corresponding value using the key. Maps store key-value pairs as elements within the container, with each element containing a key and a value. To find the value corresponding to a specific key, simply use that key as the parameter to call the map's lookup function.

[0152] In this embodiment, the cluster planning path corresponding to the cluster of unmanned driving equipment is stored using vector storage, and the location information of each target inspection key point and the unmanned driving equipment corresponding to each target inspection key point are stored using map storage. Figure 15 This application provides a schematic diagram of the structure of a cluster planning path storage format in a planning path template library; for example... Figure 15 As shown, in this embodiment, a vector is used as the outer storage format to store the cluster planning path corresponding to the unmanned driving equipment cluster, including the target planning path corresponding to each unmanned driving equipment; a map is used as the inner storage format to store the location information of each target inspection key point and the equipment information of the corresponding unmanned driving equipment, such as the equipment serial number of the unmanned driving equipment (e.g., boat 1, boat 2, etc.); therefore, the equipment information of the corresponding unmanned driving equipment can be queried through the location information of the inspection key point.

[0153] As can be seen, storing cluster planning paths according to the method of this embodiment can simplify the subsequent search process, thereby improving the convenience of the cluster planning path matching process of the autonomous driving equipment cluster.

[0154] This application also provides a cluster planning path matching method for an unmanned vehicle cluster, which can be executed by the processor of the terminal device when running the corresponding computer program.

[0155] Figure 16 The flowchart illustrates a cluster planning path matching method for an unmanned vehicle cluster provided in this embodiment. For ease of explanation, only the parts relevant to this embodiment are shown. The method provided in this embodiment includes the following steps:

[0156] S1610: Obtain the number of each unmanned driving device to be matched and the set of initial positions to be matched in the cluster of unmanned driving devices to be matched; wherein, the set of initial positions to be matched includes the initial positions to be matched corresponding to each unmanned driving device to be matched.

[0157] Among them, the cluster of autonomous driving equipment to be matched refers to the cluster of autonomous driving equipment whose corresponding cluster planning path needs to be determined; the number of equipment refers to the number of autonomous driving equipment to be matched in the cluster of autonomous driving equipment to be matched; the set of initial positions to be matched includes the initial positions to be matched corresponding to each autonomous driving equipment to be matched; the initial positions to be matched are the positions of each autonomous driving equipment to be matched in the cluster of autonomous driving equipment to be matched, which are distributed in the preset task area.

[0158] Understandably, in actual operation, the number of autonomous driving equipment clusters may increase or decrease; different numbers of equipment or different initial sets of locations to be matched will correspond to different autonomous driving equipment clusters to be matched. Therefore, it is necessary to obtain the number of autonomous driving equipment to be matched and the initial set of locations to be matched for each autonomous driving equipment in the current autonomous driving equipment cluster to be matched.

[0159] S1620: Determine the set of key points to be matched based on the initial set of positions to be matched; the set of key points to be matched includes the key points to be matched corresponding to each unmanned driving device to be matched.

[0160] Specifically, after determining the initial set of positions to be matched, the key points to be matched and inspected corresponding to each initial position are determined, thus obtaining the set of key points to be matched and inspected.

[0161] More specifically, for each initial position to be matched, the nearest patrol keypoint to that initial position is determined as the patrol keypoint corresponding to that initial position. The number of determined patrol keypoints is the same as the number of initial positions to be matched, and each patrol keypoint corresponds one-to-one with each initial position to be matched; therefore, if a patrol keypoint has already been determined to match an initial position to be matched, other initial positions to be matched can only be matched with other nearest patrol keypoints. For example... Figure 17 The diagram shown is a schematic diagram of determining a set of key points to be inspected based on a set of initial positions to be matched, according to an embodiment of this application. Figure 17 The set of key points to be matched and inspected was determined to be (1, 6, 7, 25, 31).

[0162] S1630: Based on the number of devices and the set of key points to be matched for inspection, determine the matching cluster planning path corresponding to the unmanned driving device cluster to be matched in the planning path template library; the planning path template library is obtained using any planning path template library update method; the matching cluster planning path includes the matching planning path corresponding to each unmanned driving device to be matched.

[0163] Specifically, planning path templates are sequentially determined from the planning path template library, and it is determined whether the planning path template matches the cluster of autonomous vehicles to be matched based on the set of key points to be matched. If they match, the matched planning path template is determined as the matching cluster planning path corresponding to the cluster of autonomous vehicles to be matched. The planning path template includes the planning path corresponding to each autonomous vehicle, and the corresponding matching cluster planning path includes the matching planning path corresponding to each autonomous vehicle to be matched. If the planning path template does not match the cluster of autonomous vehicles to be matched, the next planning path template is determined from the planning path template library.

[0164] It should be noted that after determining the initial set of locations to be matched, a subset of planning path templates corresponding to the number of each autonomous driving device in the autonomous driving device cluster to be matched can be selected from the planning path template library. Then, the matching cluster planning path corresponding to the autonomous driving device cluster to be matched can be determined from the planning path template subset.

[0165] This embodiment provides a cluster planning path matching method for an unmanned vehicle cluster. Because updating the planning path template library according to the update method of the planning path template library can improve the efficiency of updating the planning path template library while reducing the matching redundancy of the planning path template library; therefore, based on the planning path template library, the matching cluster planning path corresponding to the unmanned vehicle cluster to be matched can be determined in the planning path template library according to the number of devices and the set of key points to be matched, thus efficiently and conveniently determining the matching cluster planning path that matches the unmanned vehicle cluster to be matched.

[0166] Based on the above embodiments, this embodiment further explains and optimizes the technical solution. Specifically, in this embodiment, the matching cluster planning path corresponding to the unmanned driving equipment cluster to be matched is determined in the planning path template library according to the number of devices and the set of key points to be matched for inspection, including:

[0167] For each planning path template in the planning path template library, the number of unmanned vehicles corresponding to the set of key points to be matched in the planning path template is counted; the planning path template with a count equal to the number of vehicles is determined as the matching cluster planning path with the cluster of unmanned vehicles to be matched.

[0168] Specifically, for each planned path template, the following steps are performed: determine the autonomous driving equipment corresponding to the set of key inspection points to be matched in the planned path template, and count the number of autonomous driving equipment; determine whether the count is equal to the number of equipment; if it is equal to the number of equipment, then the planned path template is determined as the matching cluster planned path corresponding to the cluster of autonomous driving equipment to be matched; if the count is less than the number of equipment, it means that the planned path template does not match the cluster of autonomous driving equipment to be matched, so the next planned path template is determined from the planned path template library for matching.

[0169] Figure 18 This is a schematic diagram illustrating the process of determining the matching cluster planning path corresponding to the unmanned vehicle equipment cluster to be matched in the planning path template library based on the number of devices and the set of key points to be matched for inspection, as provided in this application embodiment. Figure 18 As shown, for this planned path template, the device serial number of the autonomous driving equipment corresponding to each key point in the set of key points to be matched (1,6,7,25,31) is determined sequentially, combined with... Figure 18 It can be seen that the correspondence between the key points to be matched and the unmanned driving equipment is: 1-boat 1, 6-boat 2, 7-boat 3, 25-boat 4, 31-boat 5. The number of different equipment serial numbers is determined to be 5, that is, the statistical number of unmanned driving equipment is determined to be 5. The statistical number is equal to the number of equipment. The planned path template is determined as the matching cluster planning path corresponding to the cluster of unmanned driving equipment to be matched.

[0170] As can be seen, the method of this embodiment can efficiently and conveniently determine the matching cluster planning path corresponding to the cluster of unmanned driving equipment to be matched.

[0171] Based on the above embodiments, this embodiment further explains and optimizes the technical solution. Specifically, in this embodiment, a cluster planning path matching method for an unmanned driving equipment cluster further includes:

[0172] If there is no planning path template whose statistical quantity is equal to the number of devices, then the set of candidate planning path templates corresponding to the maximum statistical quantity is determined; for each candidate planning path template in the set of candidate planning path templates, the total candidate distance between each initial position to be matched and each inspection key point is determined; the candidate planning path template corresponding to the minimum value of the total candidate distance is determined as the matching cluster planning path corresponding to the unmanned driving equipment cluster to be matched.

[0173] Specifically, if the number of planned path templates in the planned path template library is not equal to the number of devices, that is, there is no planned path template in the planned path template library with a number equal to the number of devices, then the candidate planned path template set corresponding to the maximum number of planned path templates is determined. The candidate planned path template set includes multiple candidate planned path templates. For each candidate planned path template, the total candidate distance between each initial position to be matched and each inspection key point is determined. The size relationship of each candidate total distance value is compared, the minimum value of the candidate total distance value is determined, and the candidate planned path template corresponding to the minimum value of the candidate total distance value is determined as the matching cluster planned path corresponding to the unmanned driving equipment cluster to be matched.

[0174] As can be seen, the method of this embodiment can efficiently and conveniently determine the planning path template corresponding to the cluster of unmanned driving equipment to be matched even when there is no planning path template with the same number of statistical items as the number of equipment.

[0175] The present application provides a method for covering and searching a cluster of unmanned vehicles, which can be executed by the processor of a terminal device when running a corresponding computer program.

[0176] Figure 19 The flowchart illustrates a coverage search method for an unmanned vehicle cluster provided in this application embodiment. For ease of explanation, only the parts relevant to this embodiment are shown. The method provided in this embodiment includes the following steps:

[0177] S1910: Determine the matching cluster planning path corresponding to the unmanned driving equipment cluster to be matched using the cluster planning path matching method of unmanned driving equipment cluster;

[0178] S1920: For each matching planning path in the matching cluster planning path, determine the target inspection key point in the matching planning path and the initial path corresponding to the target inspection key point; wherein, the initial path is the path from the initial position to be matched to the path corresponding to the target inspection key point.

[0179] The target inspection key point is the inspection key point in the matching planning path that is closest to the initial position to be matched. Specifically, after determining the matching cluster planning path corresponding to the cluster of autonomous vehicles to be matched, the matching planning path corresponding to each autonomous vehicle to be matched is determined; for each matching planning path, the straight-line distance between the initial position to be matched and each inspection key point in the matching planning path is calculated, and the inspection key point with the closest distance is determined as the target inspection key point; the initial path is determined based on the driving path between the initial position to be matched and the target inspection key point.

[0180] S1930: If there are obstacles on the initial path, the target inspection key points are re-determined based on the matching planned path, and the initial path is updated according to the updated target inspection key points so that the unmanned vehicle to be matched can perform coverage search according to the initial path and the matching planned path.

[0181] Specifically, after determining the initial path, it is determined whether there are obstacles on the initial path. Specifically, the regional properties of each task area grid on the initial path are determined. If the regional properties of all task area grids are drivable areas, it means there are no obstacles on the initial path. The autonomous vehicle to be matched first travels along the initial path to the target inspection key point, i.e., reaches the matching planned path, and then performs coverage search according to the matching planned path. If the regional properties of any task area grid are obstacles, it means there are obstacles on the initial path. Therefore, the matching planned path is adaptively adjusted, the target inspection key point is re-determined from the matching planned path, and then the initial path is updated according to the updated target inspection key point. The autonomous vehicle to be matched then travels along the updated initial path to the target inspection key point, i.e., reaches the matching planned path, and then performs coverage search according to the matching planned path.

[0182] The process of redetermining the target inspection key point from the matching planning path includes: using the Dijkstra algorithm to find the shortest path through traversal, and sequentially determining the driving path between the inspection key point in the matching planning path and the initial position to be matched; if there is an obstacle in the driving path, assign a large value to the length of the driving path to avoid the obstacle; and determining the inspection key point corresponding to the minimum value in each driving path as the target inspection key point.

[0183] The present application provides a method for covering and searching a cluster of unmanned vehicles. Since the process of determining the matching cluster planning path corresponding to the unmanned vehicle cluster to be matched according to the cluster planning path matching method of the unmanned vehicle cluster is more efficient and convenient, this method can efficiently and conveniently achieve coverage and searching of unmanned vehicle clusters.

[0184] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0185] This application embodiment also provides an apparatus for updating a route planning template library. The apparatus for updating the route planning template library in this embodiment includes:

[0186] The path determination module is used to divide the region and plan the path according to the preset task area and the initial position set to obtain the cluster planning path corresponding to the autonomous driving equipment cluster; wherein, the initial position set is the initial position corresponding to each autonomous driving equipment in the autonomous driving equipment cluster; the cluster planning path includes the target planning path corresponding to each autonomous driving equipment.

[0187] The parameter determination module is used to determine the cluster planning parameters corresponding to the cluster planning path based on the location information of the key points of the target inspection in each target planning path.

[0188] The template update module is used to add the cluster planning path to the planning path template library when it is determined from the cluster planning parameters that there is no planning path template in the planning path template library that is the same as the cluster planning path.

[0189] The updating device for a planning path template library provided in this application embodiment has the same beneficial effects as the updating method for a planning path template library described above.

[0190] This application embodiment also provides a cluster planning path matching device for an unmanned vehicle cluster. The cluster planning path matching device for the unmanned vehicle cluster in this embodiment includes:

[0191] The matching information acquisition module is used to acquire the number of each unmanned driving device to be matched and the set of initial matching positions in the unmanned driving device cluster to be matched; wherein, the set of initial matching positions is the initial matching position corresponding to each unmanned driving device to be matched.

[0192] The key point determination module is used to determine the set of key points to be matched for inspection based on the initial set of positions to be matched; the set of key points to be matched for inspection includes key points to be matched corresponding to each unmanned driving device to be matched.

[0193] The cluster matching module is used to determine the matching cluster planning path corresponding to the unmanned vehicle cluster to be matched in the planning path template library based on the number of devices and the set of key points to be matched. The planning path template library is obtained using the updating device of any planning path template library. The matching cluster planning path includes the matching planning path corresponding to each unmanned vehicle to be matched.

[0194] The cluster planning path matching device for an unmanned vehicle cluster provided in this application embodiment has the same beneficial effects as the aforementioned cluster planning path matching method for an unmanned vehicle cluster.

[0195] This application embodiment also provides a coverage search device for an unmanned vehicle cluster. The coverage search device for the unmanned vehicle cluster in this embodiment includes:

[0196] The target data acquisition module is used to determine the matching cluster planning path corresponding to the unmanned driving equipment cluster to be matched by using the cluster planning path matching device of any unmanned driving equipment cluster.

[0197] The path determination module is used to determine the target inspection key point and the initial path corresponding to the target inspection key point for each matching planning path in the matching cluster planning path; wherein, the initial path is the path from the initial position to be matched to the target inspection key point.

[0198] The obstacle avoidance module is used to redetermine the target inspection key points based on the matching planned path if there are obstacles on the initial path, and update the initial path according to the updated target inspection key points so that the unmanned driving equipment to be matched can perform coverage search according to the initial path and the matching planned path.

[0199] The coverage search device for an unmanned vehicle cluster provided in this application embodiment has the same beneficial effects as the coverage search method for an unmanned vehicle cluster described above.

[0200] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0201] This application also provides a terminal device. The terminal device of this embodiment includes a memory, a processor, and a computer program stored in the memory and executable on the processor; when the processor executes the computer program, it implements the steps in the embodiments of the above-mentioned methods for updating the planning path template library, the cluster planning path matching method for an unmanned vehicle cluster, or the coverage search method for an unmanned vehicle cluster; or when the processor executes the computer program, it implements the functions of each module / unit in the above-mentioned device embodiments.

[0202] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps in the above-described method embodiments; including a method for updating a planned path template library, a method for matching a cluster planned path of an unmanned driving equipment cluster, and a method for covering and searching an unmanned driving equipment cluster.

[0203] The computer-readable storage medium provided in this application embodiment has the same beneficial effects as the above-described method for updating a planning path template library, a method for matching a cluster planning path of an unmanned vehicle cluster, or a method for covering and searching an unmanned vehicle cluster.

[0204] The implementation of all or part of the processes in the above embodiments can be accomplished by a computer program instructing related hardware.

[0205] In the above embodiments, the descriptions of each embodiment have different focuses. Parts not described in detail or in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Those skilled in the art will recognize that the devices and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware.

[0206] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method of updating a library of path templates, characterized by, The method comprises: According to the preset task area and the initial position set, the area is divided and the path is planned, and the cluster planning path corresponding to the unmanned device cluster is obtained; wherein the initial position set is the initial position corresponding to each unmanned device in the unmanned device cluster; the cluster planning path includes the target planning path corresponding to each unmanned device; it comprises: determining a set of patrol key points corresponding to the preset task area of the unmanned device cluster and a patrol coverage area corresponding to each patrol key point in the set of patrol key points; according to the initial position set, the search area combination is iteratively constructed, and the target search area combination is determined according to the area division score of each search area combination; wherein the target search area combination includes a target search area corresponding to each unmanned device, and the target search area includes at least one patrol coverage area; for each unmanned device, the planning path is iteratively constructed based on the target search area, and the target planning path of the unmanned device is determined according to the path planning score of each planning path; According to the position information of the target patrol key point in each target planning path, the cluster planning parameter corresponding to the cluster planning path is determined; wherein the cluster planning parameter includes average product value and average sum value; When it is determined according to the cluster planning parameter that there is no planning path template identical to the cluster planning path in the planning path template library, the cluster planning path is added to the planning path template library.

2. The method of claim 1, wherein, The cluster planning parameter corresponding to the cluster planning path is determined according to the position information of the target patrol key point in each target planning path, comprising: Determine the position information of the target patrol key point of each target planning path in the cluster planning path; For each target planning path, determine the coordinate calculation value corresponding to the target planning path; According to the coordinate calculation value corresponding to each target planning path, the cluster planning parameter corresponding to the cluster planning path is determined.

3. The method of claim 1, wherein, The method further comprises: If the number of cluster planning paths in the planning path template library reaches a preset number threshold, and / or the time length of creating the planning path template library reaches a preset time length threshold, output the planning path template library.

4. The method according to any one of claims 1 to 3, characterized in that, The method further comprises: The cluster planning path corresponding to the unmanned device cluster is stored by using vector storage method; The position information of each target patrol key point and the unmanned device corresponding to each target patrol key point are stored by using map storage method.

5. A cluster planning path matching method of a cluster of unmanned devices, characterized by, The method comprises: Obtain the device quantity of each to-be-matched unmanned device in the to-be-matched unmanned device cluster and the to-be-matched initial position set; wherein the to-be-matched initial position set is the to-be-matched initial position corresponding to each to-be-matched unmanned device; Determine a set of to-be-matched patrol key points according to the to-be-matched initial position set; the set of to-be-matched patrol key points includes to-be-matched patrol key points corresponding to each to-be-matched unmanned device; determining a matching cluster planning path corresponding to the to-be-matched unmanned device cluster from the planning path template library according to the device quantity and the set of to-be-matched patrol key points; the planning path template library is obtained by using the updating method of the planning path template library according to any one of claims 1 to 4; the matching cluster planning path comprises a matching planning path corresponding to each to-be-matched unmanned device.

6. The method of claim 5, wherein, The determining a matching cluster planning path corresponding to the to-be-matched unmanned device cluster from the planning path template library according to the device quantity and the set of to-be-matched patrol key points comprises: counting the statistical quantity of the unmanned device corresponding to the set of to-be-matched patrol key points in each planning path template in the planning path template library; determining the planning path template with the statistical quantity equal to the device quantity as the matching cluster planning path of the to-be-matched unmanned device cluster.

7. The method of claim 6, wherein, The method further comprises: if there is no planning path template with the statistical quantity equal to the device quantity, determining a set of candidate planning path templates corresponding to the maximum statistical quantity; for each candidate planning path template in the set of candidate planning path templates, determining a candidate distance total value of each to-be-matched initial position and each patrol key point; determining the candidate planning path template corresponding to the minimum candidate distance total value as the matching cluster planning path corresponding to the to-be-matched unmanned device cluster.

8. A method for coverage search of a swarm of unmanned devices, characterized in that, The method comprises: determining a matching cluster planning path corresponding to the to-be-matched unmanned device cluster by using the cluster planning path matching method of the unmanned device cluster according to any one of claims 5 to 7; for each matching planning path in the matching cluster planning path, determining a target patrol key point in the matching planning path and an initial path corresponding to the target patrol key point; wherein the initial path is a path from the to-be-matched initial position to the target patrol key point; if there is an obstacle on the initial path, determining a target patrol key point again based on the matching planning path, and updating the initial path according to the updated target patrol key point, so that the to-be-matched unmanned device performs coverage search according to the initial path and the matching planning path.

9. A terminal device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 4 or claims 5 to 7 or claim 8.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. The computer program is executed by the processor to implement the steps of the method according to any one of claims 1 to 4 or claims 5 to 7 or claim 8.

Images