Autonomous fleet scheduling method and apparatus and electronic device

Abstract

An autonomous fleet scheduling method and apparatus, and an electronic device. The method comprises: carrying out global path planning on an autonomous fleet in a target area to obtain a global path planning result (S101); on the basis of travelling information of a task vehicle and reference information corresponding to a global task path, determining a local task path of the task vehicle (S102); and on the basis of attribute information of the local task path, optimizing the local task path to obtain optimized local task path information (S103). According to the method, unified global path planning on the autonomous fleet can be achieved; on the basis of real-time travelling information of the vehicle and by taking various external factors into consideration comprehensively, the local task path of the vehicle in the fleet in the travelling process can be planned, and reverse local optimization is carried out, so that the overall efficiency and scheduling accuracy of the autonomous fleet can be effectively improved while ensuring the safety; and the method can be widely popularized and applied to various complex scenarios.

Description

Unmanned vehicle fleet dispatching method, device and electronic equipment

[0001] Cross-references

[0002] This specification claims priority to the Chinese application with application number 202310898730.7 filed on July 21, 2023, entitled “A method, device and electronic device for dispatching an unmanned vehicle fleet”. The entire contents of the above application are incorporated into this specification by reference. Technical Field

[0003] This specification relates to the field of unmanned driving, and more specifically, to a method, device, and electronic equipment for dispatching an unmanned driving fleet. Background Art

[0004] With the development of autonomous driving technology, autonomous vehicle fleets are playing a vital role in various transportation scenarios. For example, the dispatch of autonomous port fleets is an essential component of the system. By planning routes and controlling the fleet's behavior, the system ensures efficient, stable, and safe operation.

[0005] When scheduling autonomous vehicle fleets, several common scheduling solutions exist: 1. Real-time control of the throttle, brakes, and other functions of all vehicles relies on significant computing power and cannot handle the general port road network. This approach is only suitable for fully automated, isolated terminals and cannot be extended to existing ports. Furthermore, magnetic nails are often required to ensure real-time position confirmation, which results in high construction and maintenance costs. 2. Independent planning and control of each vehicle often leads to deadlocks due to inter-vehicle interactions. Using traffic light assistance and physical isolation further restricts mixed traffic at the port, reducing operational efficiency. 3. Unified route planning and behavior control for the autonomous vehicle fleet, with vehicles solely responsible for positioning and execution. This approach fails to consider either local autonomous vehicle planning or the impact of external vehicles, resulting in a lack of intelligent control, incapable of tackling mixed traffic at ports, and unsuitable for general adoption.

[0006] Summary of the Invention

[0007] In view of this, this specification provides a method, device and electronic equipment for dispatching an unmanned vehicle fleet, which can not only perform unified global path planning for the unmanned vehicle fleet, but also combine the real-time driving information of the vehicle, comprehensively consider multiple external factors, plan the local task path of the fleet vehicle during driving, and perform reverse local optimization. It can effectively improve the overall efficiency and dispatching accuracy of the unmanned vehicle fleet while ensuring safety, and can be promoted and applied to a variety of complex scenarios.

[0008] An embodiment of the present specification provides a method for dispatching an unmanned vehicle fleet, the method comprising: performing global path planning on an unmanned vehicle fleet in a target area to obtain a global path planning result, wherein the global path planning result includes a global task path of a task vehicle in the unmanned vehicle fleet; determining a local task path of the task vehicle based on driving information corresponding to the task vehicle and reference information corresponding to the global task path; and optimizing the local task path based on attribute information of the local task path to obtain optimized local task path information.

[0009] In some embodiments, global path planning is performed on the unmanned driving fleet in the target area to obtain a global path planning result, including: determining a new vehicle in the unmanned driving fleet in response to a new task; determining the starting point and the end point of the new vehicle based on the new task; performing the global path planning based on the existing task paths of the existing task vehicles in the unmanned driving fleet and the starting point and the end point of the new vehicle to obtain the global path planning result.

[0010] In some embodiments, global path planning is performed on the unmanned vehicle fleet in the target area to obtain a global path planning result, including: constructing a directed road network graph corresponding to the target area, wherein, in the directed road network graph, nodes represent intersections, edges represent roads, and the weights of the edges are related to the road length and / or intersection capacity; based on the directed road network graph and the starting and ending points of the task vehicles, the global path planning is performed to obtain the global path planning result.

[0011] In some embodiments, the local task path of the task vehicle is determined based on the driving information corresponding to the task vehicle and the reference information corresponding to the global task path, including: determining the trajectory of the local task path based on obstacle information and traffic rule information; determining the speed of the local task path based on speed parameters; and determining the length of the local task path based on safety parameters and the speed parameters.

[0012] In some embodiments, the attribute information of the local task path includes at least one of the following: the trajectory of the local task path, the speed of the local task path, the length of the local task path, the priority of the local task path, and the obstacle information corresponding to the local task path.

[0013] In some embodiments, optimizing the local task path based on the attribute information of the local task path includes: optimizing the local task path based on the attribute information and an objective function, wherein the objective function is determined based on the similarity of the local task paths before and after optimization and the correlation between different local task paths, and the correlation represents the probability of collision of the task vehicles corresponding to the different local task paths.

[0014] In some embodiments, the correlation between the different local task paths is determined in the following manner: based on the attribute information corresponding to the different local task paths, the time difference between the task vehicles corresponding to the different local task paths arriving at the path intersection is determined; based on the time difference, the correlation between the different local task paths is determined.

[0015] In some embodiments, the optimized local task path information includes the optimized local task path and / or a traffic strategy of a task vehicle corresponding to the local task path.

[0016] In some embodiments, the embodiments of this specification also provide an unmanned vehicle fleet scheduling device, which includes: a global planning module, which is used to perform global path planning on the unmanned vehicle fleet in the target area to obtain a global path planning result, wherein the global path planning result includes the global task path of the task vehicle in the unmanned vehicle fleet; a local planning module, which is used to determine the local task path of the task vehicle based on the driving information corresponding to the task vehicle and the reference information corresponding to the global task path; an optimization module, which is used to optimize the local task path based on the attribute information of the local task path to obtain optimized local task path information.

[0017] In some embodiments, the embodiments of this specification also provide an electronic device, which includes: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the unmanned vehicle fleet scheduling method are performed.

[0018] The unmanned vehicle fleet scheduling method described in the embodiments of this specification not only enables unified global route planning for the unmanned vehicle fleet, but also integrates real-time vehicle driving information and comprehensively considers various external factors to plan local task routes for the fleet vehicles during their travels, and performs reverse local optimization. This method effectively improves the overall efficiency and scheduling accuracy of the unmanned vehicle fleet while ensuring safety, without requiring additional traffic lights or isolation equipment. It is rapidly replicable and can be widely applied to a variety of complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of this specification and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without any creative work.

[0020] FIG1 shows a flow chart of a method for dispatching an unmanned vehicle fleet according to an embodiment of this specification;

[0021] FIG2 shows a flowchart of global path planning for an unmanned vehicle fleet in a target area according to an embodiment of this specification;

[0022] FIG3 is a schematic diagram showing a global path planning result according to an embodiment of this specification;

[0023] FIG4 shows a flowchart of performing local path planning for a task vehicle according to an embodiment of this specification;

[0024] FIG5 is a schematic diagram showing a local path planning according to an embodiment of this specification;

[0025] FIG6 is a schematic diagram showing a local task path of a task vehicle according to an embodiment of this specification;

[0026] FIG7 shows a flowchart of local task path optimization according to an embodiment of this specification;

[0027] FIG8 is a schematic diagram showing the time difference between task vehicles corresponding to different local task paths arriving at a path intersection point according to an embodiment of this specification;

[0028] Figures 9 to 12 show schematic diagrams of the local task path optimization process;

[0029] FIG13 shows a schematic structural diagram of an unmanned vehicle fleet dispatching device according to an embodiment of this specification;

[0030] FIG14 shows a schematic structural diagram of an electronic device according to an embodiment of this specification. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the embodiments of this specification clearer, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. It should be understood that the drawings in this specification are only for the purpose of illustration and description and are not used to limit the scope of protection of this specification. In addition, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this specification illustrate the operations implemented according to some embodiments of this specification. It should be understood that the operations of the flowchart can be implemented out of sequence, and steps without logical context can be reversed or implemented simultaneously. In addition, those skilled in the art, under the guidance of the contents of this specification, can add one or more other operations to the flowchart, or remove one or more operations from the flowchart.

[0032] In addition, the described embodiments are only some of the embodiments of this specification, rather than all of the embodiments. The components of the embodiments of this specification generally described and shown in the drawings herein can be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of this specification provided in the drawings is not intended to limit the scope of the claimed specification, but merely represents selected embodiments of this specification. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this specification.

[0033] It should be noted that the term "comprising" will be used in the embodiments of this specification to indicate the existence of the features claimed thereafter, but does not exclude the addition of other features.

[0034] The embodiments of this specification provide a method, device and electronic device for dispatching an unmanned vehicle fleet. The unmanned vehicle fleet dispatching method includes: performing global path planning on the unmanned vehicle fleet in the target area to obtain a global path planning result, wherein the global path planning result includes the global task path of the task vehicle in the unmanned vehicle fleet; determining the local task path of the task vehicle based on the driving information corresponding to the task vehicle and the reference information corresponding to the global task path; optimizing the local task path based on the attribute information of the local task path to obtain the optimized local task path information. According to the unmanned vehicle fleet dispatching method described in the embodiments of this specification, it is possible to perform unified global path planning for the unmanned vehicle fleet, and it is also possible to combine the real-time driving information of the vehicle, comprehensively consider a variety of external factors, plan the local task path of the vehicle fleet during its driving process, and perform reverse local optimization. It can effectively improve the overall efficiency and dispatching accuracy of the unmanned vehicle fleet while ensuring safety, and does not require additional installation of traffic lights, isolation equipment, etc., and can be promoted and applied to a variety of complex scenarios.

[0035] Figure 1 shows a flow chart of a method for dispatching a driverless vehicle fleet according to an embodiment of this specification. In some embodiments, the method for dispatching a driverless vehicle fleet can be executed by a dispatching platform, such as a server, terminal device, processing device, electronic device, etc.

[0036] S101. Perform global path planning on the unmanned driving fleet in the target area to obtain a global path planning result, wherein the global path planning result includes a global task path of a task vehicle in the unmanned driving fleet.

[0037] The target area can refer to the place or area where the unmanned vehicle fleet performs the mission, such as ports, docks, logistics parks, bulk cargo distribution centers, etc.

[0038] An unmanned vehicle fleet can include multiple mission vehicles, each of which can interact with the dispatching platform. For example, a mission vehicle can receive control information from the dispatching platform and perform tasks accordingly. Another example is that a mission vehicle can collect driving information (e.g., positioning information, obstacle information, etc.) while driving and transmit this information to the dispatching platform.

[0039] The global task path of the task vehicle may refer to the global planned path from the starting point to the end point. For example, the starting point and / or end point may include but are not limited to quay cranes, yard cranes, forklifts, buffer zones, maintenance areas, charging areas, parking lots, etc. In the embodiments of this specification, the selection of the starting point and end point is decoupled from the work task. For example, assuming that a certain work task is to load the first cargo and the second cargo at point A, transport the first cargo to point B, and transport the second cargo to point C, the work task is decoupled, and points A and B are used as a set of starting points and end points, and points B and C are used as a set of starting points and end points, and corresponding global path planning is performed. By decoupling the starting point and end point from the work task, it can be ensured that the planning of the global path is the minimum atomic operation and will not be affected by the work task.

[0040] In some embodiments, after determining the starting and ending points of each task vehicle, the scheduling platform can perform global path planning. In some embodiments, the scheduling platform can perform global path planning based on a multi-source and multi-destination group path planning algorithm (e.g., the Floyd algorithm). In some embodiments, the scheduling platform can construct a directed road network graph corresponding to the target area and perform global path planning for the unmanned vehicle fleet based on the directed road network graph. For more details, please refer to Figures 2 and 3 and their descriptions, which will not be repeated here.

[0041] In some embodiments, during the execution of a task by an unmanned vehicle fleet, as tasks are executed, completed, and new tasks are initiated, the status of the task vehicles in the unmanned vehicle fleet is in a dynamic change, for example, the task execution state, the idle state, etc. Accordingly, in some embodiments, in response to a new task, the scheduling platform can determine the new vehicle in the unmanned vehicle fleet; and based on the new task, determine the starting point and end point of the new vehicle; and then perform global path planning based on the existing task paths of the existing task vehicles in the unmanned vehicle fleet and the starting point and end point of the new vehicle to obtain a global path planning result. In this specification, "new vehicle" can be understood as a task vehicle in an idle state or a task vehicle newly added to the current unmanned vehicle fleet; "existing task vehicle" can be understood as a task vehicle in the current unmanned vehicle fleet that is in the task execution state.

[0042] In some embodiments, when performing global path planning based on the existing mission paths of existing mission vehicles in the driverless fleet and the starting and ending points of newly added vehicles, the scheduling platform can keep the existing mission paths of the existing mission vehicles unchanged, and plan the mission paths of the newly added vehicles based on the directed road network graph and the starting and ending points of the newly added vehicles. In this case, the global path planning result includes the existing mission paths of the existing mission vehicles and the newly added mission paths of the newly added vehicles. In some embodiments, when planning the mission paths of newly added vehicles, it is necessary to consider the probability of the mission paths of the newly added vehicles colliding with the existing mission paths, and give priority to planning the new mission paths that have a lower probability of colliding with the existing mission paths. By keeping the existing mission paths unchanged, the overall amount of computation can be reduced and planning efficiency can be improved.

[0043] In some embodiments, when performing global path planning based on the existing mission paths of existing mission vehicles in the unmanned vehicle fleet and the starting and ending points of newly added vehicles, the scheduling platform can perform global path planning based on the real-time position of the existing mission vehicles, the starting and ending points of the existing mission paths, and the starting and ending points of the newly added vehicles to obtain a global path planning result. In this case, it is necessary to comprehensively consider the existing mission paths of the existing mission vehicles and the starting and ending points of the newly added vehicles, re-perform overall global planning, and determine the globally optimal global path planning result. By performing overall global planning for the existing mission vehicles and the newly added vehicles, the globally optimal global path planning result can be determined, thereby improving the accuracy of the global planning.

[0044] In some embodiments, the scheduling platform may comprehensively consider the amount of computation and accuracy, keep some existing task paths unchanged, and make overall plans for some existing task vehicles and newly added vehicles.

[0045] S102: Determine a local task path of the task vehicle based on driving information corresponding to the task vehicle and reference information corresponding to the global task path.

[0046] The driving information corresponding to the task vehicle may refer to the information detected or perceived by the task vehicle during driving. In some embodiments, the driving information corresponding to the task vehicle may include vehicle positioning information, target detection information (also referred to as "obstacle information"), etc. or any combination thereof. The vehicle positioning information may be the real-time position information detected by the positioning device of the task vehicle; the target detection information may be the relevant information (e.g., position, category, shape, size, etc.) of the targets (e.g., vehicles, pedestrians, roadside equipment, animals, stones, other task vehicles, etc.) within the perception range of the task vehicle detected by the detection equipment of the task vehicle (e.g., radar, camera, wheel speed meter, etc.).

[0047] The reference information corresponding to the global task path may refer to reference information (e.g., traffic information) related to the portion of the global task path near the current position of the task vehicle (e.g., within a preset range, within a preset distance, etc.). In some embodiments, the reference information corresponding to the global task path may include traffic rule information (e.g., left turn at the far left, right turn at the far right, waiting only after the stop line, etc.) corresponding to the portion of the global task path near the current position of the task vehicle (e.g., within a preset range, within a preset distance, etc.). In some embodiments, the reference information corresponding to the global task path may also include speed parameters, safety parameters, etc., which are used as references for planning local task paths.

[0048] The local task path of a task vehicle may refer to the local path of the task vehicle within a preset time period in the future. In some embodiments, the scheduling platform may comprehensively consider the real-time driving information corresponding to the task vehicle and reference information corresponding to the global task path (or portion thereof) to plan the local task path, thereby achieving safe and efficient intelligent control.

[0049] In some embodiments, the dispatch platform can determine the trajectory of the local task path (e.g., a trajectory that avoids obstacles and complies with traffic regulations) based on obstacle information and traffic regulation information. It can also determine the speed and length of the local task path based on speed parameters (e.g., maximum speed limit) and safety parameters (e.g., safety length and safety duration). For more details, see Figures 4-6 and their descriptions, and will not be repeated here.

[0050] In the embodiments of this specification, by comprehensively considering obstacle information, traffic rule information, speed parameters and safety parameters, local path planning can be performed to maximize the speed of the task vehicle while ensuring safety, thereby improving the overall efficiency and scheduling accuracy of the unmanned vehicle fleet.

[0051] S103: Optimize the local task path based on the attribute information of the local task path to obtain optimized local task path information.

[0052] In some embodiments, the attribute information of the local task path may include the trajectory of the local task path, the speed of the local task path, the length of the local task path, the priority of the local task path, the obstacle information corresponding to the local task path, etc. or any combination thereof.

[0053] The trajectory of the local task path can reflect the shape, direction, and direction of the local task path.

[0054] The speed of the local task path may include the speed corresponding to each trajectory point of the local task path, the global average speed, the maximum speed, the minimum speed, etc., or any combination thereof.

[0055] The length of the local task path may reflect the distance, mileage, etc. between the starting position (eg, the current position of the task vehicle) and the ending position of the local task path.

[0056] The priority of a local task path may reflect the degree of priority passage of the local task path or the corresponding task vehicle.

[0057] The obstacle information corresponding to the local task path may include obstacle information (eg, location, category, shape, size, etc.) located within a preset range of the local task path.

[0058] In some embodiments, the scheduling platform can comprehensively consider the attribute information of each local task path and optimize the traffic strategy, trajectory, length, speed, etc. of the local task path, so as to avoid collisions between task vehicles corresponding to different local task paths or avoid collisions between local task paths and nearby obstacles.

[0059] In some embodiments, the scheduling platform can optimize the local task paths based on the attribute information and the objective function of the local task paths. In some embodiments, the objective function can be determined based on the similarity of the local task paths before and after optimization and the correlation between different local task paths. In some embodiments, the correlation between different local task paths (also referred to as "correlation metric", "correlation value" or "conflict correlation") can characterize the probability of collision between task vehicles corresponding to different local task paths.

[0060] In some embodiments, the objective function may be determined based on the similarity of the local task paths before and after optimization, the correlation between different local task paths, and the priority of the local task paths.

[0061] In some embodiments, the optimized local task path information may include the optimized local task path and / or the traffic strategy (eg, passing, waiting, etc.) of the task vehicle corresponding to the local task path.

[0062] More details about local path optimization can be found in Figures 7 to 12 and their descriptions, which will not be repeated here.

[0063] In the embodiments of this specification, when optimizing local task paths, by considering the similarity of local task paths before and after optimization, the correlation between different local task paths (collision probability), the priority of local task paths, etc., the overall planning efficiency and accuracy can be further improved while ensuring safety.

[0064] In some embodiments, after obtaining optimized local task path information through local task path optimization, the dispatching platform can transmit the optimized local task path information to the task vehicle. After receiving the local task path information, the task vehicle passes or waits according to the received traffic strategy and travels along the optimized local task path.

[0065] According to the unmanned vehicle fleet scheduling method described in the embodiment of this specification, first, a global path planning is performed on the unmanned vehicle fleet in the target area to obtain a global path planning result; secondly, based on the driving information corresponding to the task vehicle and the reference information corresponding to the global task path, the local task path of the task vehicle is determined; then, based on the attribute information of the local task path, the local task path is optimized to obtain the optimized local task path information. According to the unmanned vehicle fleet scheduling method described in the embodiment of this specification, it is possible to perform unified global path planning for the unmanned vehicle fleet, and it is also possible to combine the real-time driving information of the vehicle, comprehensively consider a variety of external factors, plan the local task paths during the driving process of the fleet vehicles, and perform reverse local optimization. It can effectively improve the overall efficiency and scheduling accuracy of the unmanned vehicle fleet while ensuring safety, and does not require additional installation of traffic lights, isolation equipment, etc., and can be promoted and applied to a variety of complex scenarios.

[0066] In addition, whether it is local task path planning or further optimization of local task paths, the unmanned driving fleet scheduling method described in this specification is planned and optimized from a global level (rather than just for a single task vehicle or a single intersection), thereby improving the scheduling efficiency and accuracy of the entire unmanned driving fleet.

[0067] In some embodiments, during the task execution process, the scheduling platform can dynamically execute steps S101, S102 and S103 in a loop, thereby dynamically updating the global task path, local task path and optimized local task path information. In some embodiments, the scheduling platform can perform global path planning at a first preset time interval and perform local task path planning at a second preset time interval, wherein the first preset time interval (for example, 10 seconds) is greater than the second preset time interval (for example, 1 second). As an example only, the scheduling platform detects whether there are new tasks at the order of 10 seconds. If so, the response is used for the new tasks, performs global path planning for the unmanned vehicle fleet in the target area, and updates the global path planning results; in addition, the local task path is updated at the order of 1 second. By setting different update intervals, the overall scheduling efficiency of the unmanned vehicle fleet can be improved while ensuring safety.

[0068] It should be noted that the above description of the overall process is for illustration and purpose only and does not limit the scope of application of this specification. Those skilled in the art may make various modifications and alterations to the above process under the guidance of this specification. However, such modifications and alterations are still within the scope of this specification.

[0069] FIG2 shows a flowchart of global path planning for an unmanned vehicle fleet in a target area according to an embodiment of this specification.

[0070] S201: Construct a directed road network graph corresponding to the target area.

[0071] In some embodiments, the dispatching platform may construct a directed road network graph by treating road intersections (i.e., intersections) in the target area as nodes and roads between intersections as edges. In some embodiments, the dispatching platform may construct a directed road network graph based on a map (e.g., an electronic map) of the target area.

[0072] In some embodiments, in a directed road network graph, nodes represent intersections, edges represent roads, and edge weights are related to road length and / or intersection capacity. For example, longer roads have greater edge weights, while greater intersection capacity has smaller edge weights. In particular, when intersection capacity is low, the weight increases rapidly.

[0073] In some embodiments, the edge weights may be defined as:

[0074] G ij =exp{-T L / L r}·T c / C r (1)

[0075] Among them, G ij represents the weight of the edge; Lr Indicates the road length; T L Represents the road constraint parameter, which is used to constrain the value range; C r represents the intersection capacity; T C In some embodiments, T L and T C It can be determined based on historical data of the target area. In some embodiments, T L and T C It can be a system default value or a user-set value. In some embodiments, T L and T C It can be dynamically adjusted for different application scenarios.

[0076] S202: Based on the directed road network graph and the starting point and end point of the task vehicle, global path planning is performed to obtain a global path planning result.

[0077] In some embodiments, the scheduling platform can perform global path planning based on a multi-source, multi-destination group path planning algorithm (e.g., the Floyd algorithm). In some embodiments, the global path planning result is a global or group optimum, rather than an individual optimum.

[0078] As an example only, FIG3 shows a schematic diagram of the global path planning result described in the embodiment of this specification. As shown in FIG3 , the starting point of the task vehicle 300 is point A and the end point is point B. There are multiple paths to choose from between point A and point B, for example, a first path 301 and a second path 302, and the length of the first path 301 is shorter than the length of the second path 302. From the perspective of individual optimality, the planned path of the task vehicle 300 should be the first path 301; from the perspective of global optimality, it is necessary to comprehensively consider the path conflicts with other task vehicles, the path lengths of other task vehicles, etc., and comprehensively select the second path 302 that is globally optimal.

[0079] It should be noted that the above description of the global path planning process is for illustrative purposes only and does not limit the scope of application of this specification. Those skilled in the art may make various modifications and alterations to the above process under the guidance of this specification. However, such modifications and alterations remain within the scope of this specification.

[0080] FIG4 shows a flowchart of performing local path planning for a task vehicle according to an embodiment of this specification.

[0081] S401: Determine the trajectory of the local task path based on obstacle information and traffic rule information.

[0082] In combination with step S102, the obstacle information may be relevant information (e.g., location, category, shape, size, etc.) of targets (e.g., vehicles, pedestrians, roadside equipment, animals, stones, other task vehicles, etc.) within the perception range of the task vehicle; the traffic rule information may be relevant rules for vehicles driving on the road (e.g., left turn at the leftmost side, right turn at the rightmost side, waiting only after the stop line, etc.).

[0083] Accordingly, the trajectory of the local mission path is a trajectory that avoids obstacles near the mission vehicle and complies with traffic regulations.

[0084] As an example, Figure 5 shows a schematic diagram of local path planning as described in an embodiment of this specification. As shown in Figure 5, after global path planning, the global task path of task vehicle 501 is 502; after local task path planning, its local task path is 504, avoiding nearby obstacles 503 (e.g., pedestrians, vehicles, etc.).

[0085] S402: Determine the speed of the local task path based on the speed parameter.

[0086] In some embodiments, the scheduling platform can set the speed of the local task path based on the speed parameter. For example, the speed of each path point of the local task path is set to be the same as the speed parameter. For another example, the average speed of the local task path is set to be the same as the speed parameter. For another example, the maximum speed of the local task path is set to be the same as the speed parameter. In some embodiments, the speed parameter can be the maximum speed limit of the road. In some embodiments, the speed parameter can be the maximum speed that can be achieved historically. In some embodiments, the speed parameter can be the highest safe speed that can be matched with the task type or vehicle type. For example, the greater the load of the task vehicle, the lower the maximum safe speed. In some embodiments, the speed parameter can be a pre-set maximum speed of the vehicle.

[0087] In the embodiments of this specification, when planning a local task path, increasing the speed of the local task path as much as possible can improve the work efficiency of the unmanned vehicle fleet as much as possible.

[0088] S403: Determine the length of the local task path based on the safety parameter and the speed parameter.

[0089] In some embodiments, the scheduling platform can plan the length of the local task path by comprehensively considering safety and efficiency. In some embodiments, safety parameters may include safety length, safety duration, etc. In some embodiments, the safety length and / or safety duration can be used to ensure the safe parking and / or starting of the task vehicle.

[0090] In some embodiments, the scheduling platform may determine the length of the local task path based on the following formula:

[0091] w i =w basis +v i ·t basis (2)

[0092] Among them, w i represents the length of the local task path; w basis Indicates safety length (also known as "safety tolerance"); v i represents the speed parameter (for example, the maximum speed limit of the road); t basis In some embodiments, t basis It can be determined based on historical data of the target area. In some embodiments, t basis It can be a system default value or a user-set value. In some embodiments, t basis It can be dynamically adjusted for different application scenarios.

[0093] As an example only, Figure 6 illustrates a schematic diagram of the localized task paths of the task vehicles described in the embodiments of this specification. As shown in Figure 6 , after localized path planning for the four task vehicles, a first localized task path corresponding to the first task vehicle 601, a second localized task path corresponding to the second task vehicle 602, a third localized task path corresponding to the third task vehicle 603, and a fourth localized task path corresponding to the fourth task vehicle 604 are determined.

[0094] It should be noted that the above description of the local path planning process is for illustrative purposes only and does not limit the scope of application of this specification. Those skilled in the art may make various modifications and alterations to the above process under the guidance of this specification. However, such modifications and alterations remain within the scope of this specification.

[0095] FIG7 shows a flowchart of local task path optimization according to an embodiment of this specification.

[0096] S701. Based on attribute information corresponding to different local task paths, determine the time difference between task vehicles corresponding to different local task paths arriving at a path intersection.

[0097] In some embodiments, the scheduling platform can determine the time difference between the task vehicles corresponding to different local task paths arriving at the path intersection based on the trajectory, speed, length, real-time position of the corresponding task vehicles, etc. of different local task paths.

[0098] As an example only, FIG8 shows a schematic diagram of the time difference between the task vehicles corresponding to different local task paths reaching the path intersection point described in the embodiments of this specification. As shown in FIG8, the local task path of task vehicle 801 does not have an intersection with the local task paths of other task vehicles, that is, no collision can occur; task vehicle 802 and task vehicle 804 both arrive at the path intersection point of their corresponding local task paths 2 seconds later, so the time difference is 0 seconds; for task vehicle 803 and task vehicle 804, task vehicle 803 will arrive at the path intersection point of their corresponding local task paths 5 seconds later, and task vehicle 804 will arrive at the path intersection point 1 second later, so the time difference is 4 seconds.

[0099] S702: Determine the correlation between different local task paths based on the time difference.

[0100] In some embodiments, the greater the time difference between the task vehicles arriving at the path intersection, the lower the probability of the task vehicles colliding, and the lower the correlation between the corresponding local task paths; the smaller the time difference between the task vehicles arriving at the path intersection, the higher the probability of the task vehicles colliding, and the higher the correlation between the corresponding local task paths.

[0101] In some embodiments, the correlation between different local task paths can be defined as:

[0102] Among them, l ij Indicates the correlation between two local task paths; tcc ij It represents the time difference between the task vehicles corresponding to the two local task paths arriving at the intersection; the smaller the time difference, the better. ij The larger the time difference is, the closer it is to 0. ij Close to infinity; the larger the time difference, l ij The smaller it is, in particular, if there is no intersection between two local task paths, then l ij =0; T tcc Represents a correlation constraint parameter, used to constrain the value range. In some embodiments, T tcc It can be determined based on historical data of the target area. In some embodiments, T tcc It can be a system default value or a user-set value. In some embodiments, T tcc It can be dynamically adjusted for different application scenarios. In some embodiments, combined with formula (2), T tcc Can be used with t basis In some embodiments, the two values ​​may not be equal or may take independent values.

[0103] S703: Optimize the local task paths based on the attribute information and the objective function of the local task paths, wherein the objective function is determined based on at least the correlation between different local task paths.

[0104] In some embodiments, when optimizing local task paths, the objective function aims to maximize the similarity of local task paths before and after optimization and minimize the correlation between different local task paths, thereby achieving safe and efficient planning goals.

[0105] In some embodiments, the objective function may be defined as:

[0106] min∑ i ∑ j x i ·l ij ·x j (4)

[0107] in, and represents the local task path; x i ∈{0,1} indicates whether the i-th local task path is selected (which can be understood as the passage strategy of the local task path. For example, if a local task path is selected, its passage strategy is "pass", and if it is not selected, its passage strategy is "wait"); x i ∈{0,1} indicates whether the jth local task path is selected; l ij Represents the correlation between two local task paths.

[0108] In the embodiments of the present specification, by introducing the similarity of the local task paths before and after optimization into the objective function, the optimized local task path can still be driven as close to the preset speed parameters (for example, the maximum speed limit of the road) as possible. In addition, by introducing the correlation between different local task paths into the objective function, it can be ensured that paths with a higher probability of collision will not be selected at the same time, thereby ensuring driving safety.

[0109] In some embodiments, when optimizing local task paths, the priority of the local task paths can also be introduced into the objective function. In other words, the objective function can be determined based on the similarity of the local task paths before and after optimization, the correlation between different local task paths, and the priority of the local task paths. Details about priority can be found below.

[0110] In the embodiments of this specification, based on the similarity of local task paths before and after optimization and the correlation between different local task paths, the efficiency of local task path optimization can be further improved by introducing the priority of local task paths into the objective function.

[0111] In some embodiments, the scheduling platform can optimize and solve local task paths based on a graph theory solution.

[0112] In some embodiments, the objective function matrix can be defined as follows:

[0113] in, The matrix representation of the local task path, for example, Local task path The matrix is ​​composed of: X represents the matrix representation of whether the local task path is selected after optimization, for example, X is x i The matrix is ​​composed of: L represents the matrix representation of the correlation between local task paths (also called "correlation matrix"), for example, L is l ij The matrix formed by I c The matrix representation of the priority or confidence of the local task path (also called "priority matrix" or "confidence matrix"), for example, I c is the priority or confidence c i The diagonal matrix is ​​composed of ; λ is a preset parameter used to balance the weights of similarity and priority.

[0114] In some embodiments, the priority (or confidence) of a local task path may be related to the priority of the task performed by the corresponding task vehicle. For example, the higher the task priority, the higher the priority of the local task path. For example, the priority of the loading task is the highest, the priority of the unloading task is the second highest, and the priority of the box moving task is the lowest. In some embodiments, the priority (or confidence) of a local task path may be related to the historical waiting time of the corresponding task vehicle. For example, the longer the historical waiting time, the higher the priority; if the task vehicle has never waited in the past, the priority can even be set to 0 (i.e., the priority of the local task path of the task vehicle does not need to be considered during the solution process). In some embodiments, when determining the priority (or confidence) of a local task path, the priority of the task performed by the corresponding task vehicle and the historical waiting time may be comprehensively considered.

[0115] In some embodiments, the priority of the local task path can be defined as:

[0116] c i =f i ·exp{-T t / ||t i ||} (7)

[0117] Among them, c i represents the priority of the local task path, f iIndicates the task priority corresponding to the local task path; t i represents the waiting time of the task vehicle corresponding to the local task path; T t Represents a weight constraint parameter, which is used to constrain the weight of the waiting time. In some embodiments, T t It can be determined based on historical data of the target area. In some embodiments, T t It can be a system default value or a user-set value. In some embodiments, T t It can be dynamically adjusted for different application scenarios.

[0118] In some embodiments, based on the above, the objective function matrix can be derived based on the graph theory solution, and the optimization result of the local task path can be solved by the Jacobian matrix. As an example only, the solution process is as follows:

[0119] In some embodiments, the optimized local task path information (also referred to as "optimization result of the local task path") may include the passage strategy (eg, passage, waiting) of the task vehicle corresponding to the local task path.

[0120] As an example only, Figures 9-12 show schematic diagrams of the local task path optimization process. As shown in Figure 9, in conjunction with Figure 6, the local task path of the first task vehicle 601 does not have an intersection with the local task paths of other task vehicles, and the local task paths of the second task vehicle 602, the third task vehicle 603, and the fourth task vehicle 604 do have an intersection. If each task vehicle passes according to its own local task path, a collision may occur. Accordingly, the scheduling platform can determine the traffic strategy of the first task vehicle 601, the second task vehicle 602, the third task vehicle 603, and the fourth task vehicle 604 through the above optimization process. As shown in Figure 9, the traffic strategy at the current moment is: the first task vehicle 601 passes normally along the first local task path, and the second task vehicle 602, the third task vehicle 603, and the fourth task vehicle 604 all pass until the local stop line (for example, near the intersection of the second local task path, the third local task path, and the fourth local task path).

[0121] Furthermore, in conjunction with the above, the optimization process for the local task paths continues dynamically. As shown in Figure 10 , the traffic strategy at this moment is: the first task vehicle 601 continues to pass, the third task vehicle 603 passes through the intersection of the second, third, and fourth local task paths, and the second and fourth task vehicles 602 and 604 wait at the local stop line.

[0122] Furthermore, as shown in Figure 11, the first task vehicle 601 has driven out of the local path optimization range. The traffic strategy at this moment is: the third task vehicle 603 continues to pass, the second task vehicle 602 passes through the intersection of the second local task path and the fourth local task path, and the fourth task vehicle 604 waits at the local stop line.

[0123] Furthermore, as shown in FIG12 , the traffic strategy at this moment is: the third task vehicle 603 continues to pass, the second task vehicle 602 has passed the intersection of the second local task path and the fourth local task path, and the fourth task vehicle 604 starts to pass.

[0124] In some embodiments, the optimized local task path information (also referred to as "optimization result of the local task path") may also include the optimized local task path. That is to say, when performing local task path optimization, the scheduling platform not only considers the traffic strategy of the local task path (for example, passage, waiting), but also adjusts the local task path during the optimization process (for example, adjusting the trajectory, length, speed, etc. of the local task path). For example, during the optimization process, the scheduling platform can adjust the trajectory of the local task path (for example, changing lanes) to avoid collisions between task vehicles corresponding to different local task paths or to avoid collisions between local task paths and obstacles. For another example, during the optimization process, the scheduling platform can adjust the speed of the local task path (for example, reducing the speed) to increase the time difference between different local task paths arriving at the path intersection, and accordingly avoid collisions between corresponding task vehicles.

[0125] Accordingly, in conjunction with Formula 6 above, when optimizing local task paths based on the objective function, X represents the matrix representation of the optimized local task paths and whether they are selected, and L represents the matrix representation of the correlations between the optimized local task paths. Furthermore, in conjunction with step S702, the correlations between the optimized local task paths are determined based on the time differences between the task vehicles corresponding to the optimized local task paths arriving at the path intersection.

[0126] In some embodiments, since the correlation characterizes the probability of collision of task vehicles corresponding to different local task paths, and the local task path itself takes into account the safety length and safety time, therefore, if the local task path changes little before and after optimization, and the change in the probability of collision of task vehicles caused by the change in path trajectory before and after optimization is within the allowable error range, the local task path before optimization can still be used to determine the time difference, and then determine the correlation.

[0127] It should be noted that the above description of the local path optimization process is for illustrative purposes only and does not limit the scope of application of this specification. Those skilled in the art will be able to make various modifications and alterations to the above process under the guidance of this specification. However, such modifications and alterations remain within the scope of this specification.

[0128] Based on the same inventive concept, embodiments of this specification also provide an unmanned vehicle fleet scheduling device corresponding to the unmanned vehicle fleet scheduling method. Figure 13 shows a schematic structural diagram of the unmanned vehicle fleet scheduling device described in embodiments of this specification. As shown in Figure 13, the unmanned vehicle fleet scheduling device includes a global planning module 1301, a local planning module 1302, and an optimization module 1303.

[0129] The global planning module 1301 can be used to perform global path planning for the unmanned driving fleet in the target area to obtain a global path planning result, wherein the global path planning result includes the global task path of the task vehicles in the unmanned driving fleet.

[0130] The local planning module 1302 may be configured to determine a local task path of the task vehicle based on driving information corresponding to the task vehicle and reference information corresponding to the global task path.

[0131] The optimization module 1303 may be used to optimize the local task path based on the attribute information of the local task path to obtain optimized local task path information.

[0132] In some embodiments, when performing global path planning for an unmanned vehicle fleet in a target area, the global planning module 1301 can determine new vehicles in the unmanned vehicle fleet in response to new tasks; determine the starting and ending points of the new vehicles based on the new tasks; and perform global path planning based on the existing task paths of existing task vehicles in the unmanned vehicle fleet and the starting and ending points of the new vehicles to obtain a global path planning result.

[0133] In some embodiments, when performing global path planning for an unmanned vehicle fleet in a target area, the global planning module 1301 can construct a directed road network graph corresponding to the target area; based on the directed road network graph and the starting and ending points of the task vehicle, global path planning is performed to obtain a global path planning result.

[0134] In some embodiments, when determining the local task path of the task vehicle, the local planning module 1302 can determine the trajectory of the local task path based on obstacle information and traffic rule information; determine the speed of the local task path based on speed parameters; and determine the length of the local task path based on safety parameters and speed parameters.

[0135] In some embodiments, when optimizing a local task path, the optimization module 1303 may optimize the local task path based on attribute information and an objective function, wherein the objective function is determined based on the similarity of the local task paths before and after optimization and the correlation between different local task paths, and the correlation represents the probability of collision between task vehicles corresponding to different local task paths.

[0136] In some embodiments, the unmanned vehicle fleet scheduling device may also include a correlation determination module (not shown in the figure), which is used to determine the time difference between the task vehicles corresponding to different local task paths arriving at the path intersection based on the attribute information corresponding to different local task paths; and based on the time difference, determine the correlation between different local task paths.

[0137] The specific details can be found in other parts of this manual and will not be repeated here.

[0138] Based on the same inventive concept, an embodiment of this specification also provides an electronic device corresponding to the unmanned vehicle fleet scheduling method. Figure 14 shows a schematic diagram of the structure of the electronic device described in the embodiment of this specification. As shown in Figure 14, the electronic device 1400 includes a processor 1402, a memory 1401, and a bus. The memory 1401 stores machine-readable instructions executable by the processor 1402. When the electronic device 1400 is running, the processor 1402 communicates with the memory 1401 via the bus. When the machine-readable instructions are executed by the processor 1402, the steps of the unmanned vehicle fleet scheduling method described in this specification are executed.

[0139] Based on the same inventive concept, embodiments of this specification also provide a computer-readable storage medium corresponding to the unmanned vehicle fleet scheduling method. The computer-readable storage medium stores a computer program that, when executed by a processor, executes the steps of the unmanned vehicle fleet scheduling method described in this specification.

[0140] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the method embodiments, and will not be repeated in this specification. In the several embodiments provided in this specification, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0141] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0142] In addition, each functional unit in each embodiment of this specification may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0143] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of this specification, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, platform server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of this specification. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

[0144] The above are only specific embodiments of this specification, but the scope of protection of this specification is not limited to them. Any changes or substitutions that can be easily conceived by any person skilled in the art within the technical scope disclosed in this specification should be included in the scope of protection of this specification. Therefore, the scope of protection of this specification should be based on the scope of protection of the claims.

Claims

1. A method for dispatching an unmanned vehicle fleet, characterized in that: The method comprises: Performing global path planning on the unmanned driving fleet in the target area to obtain a global path planning result, wherein the global path planning result includes a global task path of the task vehicles in the unmanned driving fleet; Determining a local task path of the task vehicle based on the driving information corresponding to the task vehicle and the reference information corresponding to the global task path; and Based on the attribute information of the local task path, the local task path is optimized to obtain optimized local task path information.

2. The unmanned vehicle fleet dispatching method according to claim 1, characterized in that: Perform global path planning for the driverless fleet in the target area and obtain the global path planning results, including: In response to the newly added task, determining a new vehicle for the unmanned vehicle fleet; Based on the newly added task, determining the starting point and the end point of the newly added vehicle; and Based on the existing mission paths of the existing mission vehicles in the unmanned vehicle fleet and the starting point and the end point of the newly added vehicle, the global path planning is performed to obtain the global path planning result.

3. The unmanned vehicle fleet dispatching method according to claim 1, characterized in that: Perform global path planning for the driverless fleet in the target area and obtain the global path planning results, including: Constructing a directed road network graph corresponding to the target area, wherein in the directed road network graph, nodes represent intersections, edges represent roads, and weights of the edges are related to road lengths and / or intersection capacities; and Based on the directed road network graph and the starting point and the end point of the task vehicle, the global path planning is performed to obtain the global path planning result.

4. The unmanned vehicle fleet dispatching method according to claim 1, characterized in that: Determining a local task path of the task vehicle based on the driving information corresponding to the task vehicle and the reference information corresponding to the global task path includes: Determining the trajectory of the local task path based on obstacle information and traffic rule information; Based on the speed parameter, determining the speed of the local task path; and Based on the safety parameter and the speed parameter, the length of the local task path is determined.

5. The unmanned vehicle fleet dispatching method according to claim 1, characterized in that: The attribute information of the local task path includes at least one of the following: the trajectory of the local task path, the speed of the local task path, the length of the local task path, the priority of the local task path, and obstacle information corresponding to the local task path.

6. The unmanned vehicle fleet dispatching method according to claim 1, characterized in that: Optimizing the local task path based on the attribute information of the local task path includes: Based on the attribute information and the objective function, the local task path is optimized, wherein the objective function is determined based on the similarity of the local task paths before and after optimization and the correlation between different local task paths, and the correlation represents the probability of collision between the task vehicles corresponding to the different local task paths.

7. The unmanned vehicle fleet dispatching method according to claim 6, characterized in that: The correlation between the different local task paths is determined in the following way: Based on the attribute information corresponding to the different local task paths, determining the time difference of the task vehicles corresponding to the different local task paths arriving at the path intersection point; and Based on the time difference, correlations between the different local task paths are determined.

8. The method for dispatching an unmanned vehicle fleet according to claim 1, characterized in that: The optimized local task path information includes the optimized local task path and / or the traffic strategy of the task vehicle corresponding to the local task path.

9. An unmanned vehicle fleet dispatching device, characterized in that: The device comprises: A global planning module, used to perform global path planning for the unmanned driving fleet in the target area to obtain a global path planning result, wherein the global path planning result includes a global task path of the task vehicles in the unmanned driving fleet; A local planning module, configured to determine a local task path of the task vehicle based on the driving information corresponding to the task vehicle and the reference information corresponding to the global task path; and The optimization module is used to optimize the local task path based on the attribute information of the local task path to obtain optimized local task path information.

10. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the unmanned vehicle fleet scheduling method as described in any one of claims 1 to 8 are performed.

Images