Unmanned aerial vehicle and unmanned vehicle cooperative path planning method based on attention mechanism

Abstract

According to the unmanned aerial vehicle and unmanned vehicle cooperative path planning method based on the attention mechanism, for a data acquisition task of a region omega, it is assumed that a UGV plans a shortest driving path according to a road point needing to be accessed. Then, a certain number of unmanned aerial vehicles with limited energy consumption are dispatched to access the cluster head, and due to the fact that the unmanned aerial vehicles need to be charged with the unmanned vehicles at a certain position, cooperative path planning of the unmanned aerial vehicles and the unmanned vehicles needs to be considered; after a path is found for the UGV, a sufficient number of vertexes are distributed on the path to serve as charging positions of UAV-UGV intersection. A vehicle routing problem with energy constraints is constructed by assuming that k UAVs and UGVs start from the same location base station to consider energy constraints, time windows allowing meeting, and some of UGV path vertexes allowing UAVs to access. The problem is solved by expressing the problem as having a VRP problem with energy constraints, time windows, and drop accesses.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) navigation technology, and in particular to a collaborative path planning method for UAVs and unmanned vehicles based on an attention mechanism. Background Art

[0002] When unmanned systems gather information, traditional drone swarms are constrained by the drones' energy consumption and limited endurance, resulting in long latency in data transmission. By providing rechargeable unmanned vehicles (UGVs) to power the unmanned aerial vehicles (UAVs), drone swarms can quickly provide reliable and cost-effective wireless communication within a designated area, completing information collection tasks within a specific area with reduced latency. To enable the UAVs to rendezvous with the drones for recharging, the drones and UGVs must collaborate, planning their flight paths and rendezvousing at appropriate locations to recharge, minimizing the time it takes for the drones to gather information.

[0003] Currently, there are many different path planning algorithms, which can be broadly categorized into graph-based algorithms, mathematical programming-based algorithms, and intelligent heuristic algorithms. Specifically, there are A* Algorithm, Rapidly Exploring Random Tree (RRT), Probabilistic Roadmap (PRM), Dynamic Programming, as well as heuristic intelligent search algorithms such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Ant Colony Optimization (Ant Colony Optimization), and Simulated Annealing.

[0004] However, these methods can provide feasible solutions in specific scenarios, but when faced with large-scale, complex-constrained path planning problems, they often have problems such as high computational complexity, lack of dynamic adjustment capabilities, difficulty in handling complex constraints, and poor coordination capabilities. Summary of the Invention

[0005] The problem to be solved by this invention is to design a collaborative path planning method for drones and unmanned vehicles based on the attention mechanism in response to the shortcomings of the existing technology. The method aims to plan a reasonable flight route for the drone and rendezvous with the unmanned vehicle at a suitable location for charging, thereby minimizing the time for drone information collection.

[0006] In a first aspect, the present invention provides a method for collaborative path planning between a drone and an unmanned vehicle based on an attention mechanism, comprising the following steps: Step 1: Determine the three-dimensional range of the target area Ω and set the flight altitude limit of the UAV; Step 2: Deploy sensors for collecting information data of the target area Ω and cluster heads for receiving and storing the information data collected by the sensors within the target area Ω. Step 3: Set the scene parameters: The uniform speed of the unmanned vehicle V_{G}; The charging rate of the autonomous vehicle C_{rate}; The uniform flight speed of the UAV is V_{A}; The energy consumption rate of the drone E_{rate}; The maximum communication distance D between the UAV and the unmanned vehicle; The maximum communication distance D_{cluster} between the UAV and the cluster head; Step 4: Unmanned vehicle route planning; Perform k-means clustering on the cluster head positions, and the obtained cluster centers are used as the waypoints that the unmanned vehicle needs to visit; Using waypoints as vertices, construct and solve a traveling salesman problem (TSP) to find the shortest path for the autonomous vehicle. The constructed traveling salesman problem TSP is: Undirected graph G'=(V',E'), where V'={0,1,...,k} is the entire vertex set, 0 is the starting vertex; E' is the edge set, representing the connection relationship between vertices; c_{ij} is the non-negative arc cost between vertices i and j; x_{ij} is a binary variable. If the autonomous vehicle travels from i to j, then x_{ij}=1, otherwise it is 0. The constraints of the traveling salesman problem TSP are: The goal is to minimize the total distance traveled by the autonomous vehicle; Ensure that the autonomous vehicle visits each vertex once; Ensure that the unmanned vehicle starts from the base station vertex and ends at the base station vertex; Make sure there is no sub-tour; Constraints on binary decision variables x_{ij}; Step 5: Collaborative path planning between the unmanned vehicle and the drone; Allocate a sufficient number of vertices on the path of the unmanned vehicle as the rendezvous and charging locations for the UAV and the unmanned vehicle; Assuming that there are k drones and unmanned vehicles starting from the same base station, a vehicle routing problem (VRP) with energy constraints is constructed, taking into account the energy limit of the drones, the time window allowed for meeting, and the constraints on the drones visiting the vertices of the unmanned vehicle's path; The constructed vehicle routing problem VRP is: The set of all unmanned vehicle waypoints A={1,2,...,m}; Pre-designated charging intersection point M={m+1,...,n}; The set of all vertices V=M∪A={0,1,2,...,m,m+1,...,n}; The set of all edges E = {(i, j) | i, j∈V, i≠j}; The constraints are: Drones and autonomous vehicles must meet at designated charging intersections; The drone’s flight path must meet energy consumption constraints; The drone’s flight path must be coordinated with the path of the unmanned vehicle to ensure timely charging at the charging intersection; Step 6: Based on the collaborative path planning of the unmanned vehicle and the drone, an attention mechanism is introduced to optimize the flight path of the drone. In step 6.1, an attention weight is assigned to each UAV, and the weight is dynamically adjusted according to the importance of the cluster head, the energy status of the UAV, and the task priority; In step 6.2, the attention mechanism is used to calculate the path selection probability of the UAV at each time step, giving priority to paths with lower energy consumption or higher task priority; Step 6.3, dynamically adjust the flight path of the UAV to ensure that it can meet the unmanned vehicle at the charging intersection before the energy is exhausted; Step 7: During the actual execution, the drone and the unmanned vehicle move along the planned path. The drone monitors energy consumption in real time during flight and selects the optimal path based on the dynamic adjustment results of the attention mechanism. The unmanned vehicle follows the planned path and waits for the drone to charge at the charging intersection. If the drone's energy is consumed too quickly, it will prioritize charging at the nearest charging junction; If the task priority changes, the UAV dynamically adjusts its path to give priority to visiting the new high-priority cluster head; During driving, the unmanned vehicle adjusts its driving speed according to the real-time status of the drone to ensure that it meets the drone at the charging intersection; Step 8: After the UAV completes the visit mission to all cluster heads, the UAV and the unmanned vehicle return to the base station according to the planned path; based on the task execution status, the optimization suggestions for the path planning are fed back.

[0007] In a second aspect, the present invention provides a collaborative path planning system for drones and unmanned vehicles based on an attention mechanism, comprising: The target area determination module is used to determine the three-dimensional range of the target area Ω and set the flight altitude limit of the UAV; Sensors are deployed in the target area Ω to collect information data of the target area Ω; Cluster head, deployed in the target area Ω, is used to receive and store information data collected by sensors; The parameter setting module is used to set the scenario parameters: the uniform driving speed of the unmanned vehicle V_{G}; the charging rate of the unmanned vehicle C_{rate}; the uniform flight speed of the drone V_{A}; the energy consumption rate of the drone E_{rate}; the maximum communication distance D between the drone and the unmanned vehicle; the maximum communication distance D_{cluster} between the drone and the cluster head; Unmanned vehicle route planning module, used for: Perform k-means clustering on the cluster head positions, and the obtained cluster centers are used as the waypoints that the unmanned vehicle needs to visit; Cluster the cluster head positions again to obtain a set of sparse waypoints; Using waypoints as vertices, construct and solve a traveling salesman problem (TSP) to find the shortest path for the autonomous vehicle. Among them, the constructed traveling salesman problem TSP is: Undirected graph G'=(V',E'), where V'={0,1,...,k} is the entire vertex set, 0 is the starting vertex; E' is the edge set, representing the connection relationship between vertices; c_{ij} is the non-negative arc cost between vertices i and j; x_{ij} is a binary variable. If the autonomous vehicle travels from i to j, then x_{ij}=1, otherwise it is 0. Among them, the constraints of the traveling salesman problem TSP are: The goal is to minimize the total distance traveled by the autonomous vehicle; Ensure that the autonomous vehicle visits each vertex once; Ensure that the unmanned vehicle starts from the base station vertex and ends at the base station vertex; Make sure there is no sub-tour; Constraints on binary decision variables x_{ij}; Unmanned vehicle and drone collaborative path planning module for: Allocate a sufficient number of vertices on the path of the unmanned vehicle as the rendezvous and charging locations for the UAV and the unmanned vehicle; Assuming that there are k drones and unmanned vehicles starting from the same base station, a vehicle routing problem (VRP) with energy constraints is constructed, taking into account the energy limit of the drones, the time window allowed for meeting, and the constraints on the drones visiting the vertices of the unmanned vehicle's path; Among them, the constructed vehicle routing problem VRP is: The set of all unmanned vehicle waypoints A={1,2,...,m}; Pre-designated charging intersection point M={m+1,...,n}; The set of all vertices V=M∪A={0,1,2,...,m,m+1,...,n}; The set of all edges E = {(i, j) | i, j∈V, i≠j}; The constraints are: Drones and autonomous vehicles must meet at designated charging intersections; The drone’s flight path must meet energy consumption constraints; The drone’s flight path must be coordinated with the path of the unmanned vehicle to ensure timely charging at the charging intersection; The attention mechanism optimization module is used to introduce the attention mechanism to optimize the flight path of the drone based on the collaborative path planning of the unmanned vehicle and the drone; Among them, the implementation method of the attention mechanism optimization module is: Assign an attention weight to each UAV, which is dynamically adjusted based on the importance of the cluster head, the energy status of the UAV, and the task priority; Use the attention mechanism to calculate the path selection probability of the drone at each time step, giving priority to paths with lower energy consumption or higher task priority; Dynamically adjust the drone's flight path to ensure it can meet the unmanned vehicle at the charging intersection before its energy is exhausted; The path execution and dynamic adjustment module is used to: During the actual execution process, the drone and the unmanned vehicle move according to the planned path; the drone monitors energy consumption in real time during flight and selects the optimal path based on the dynamic adjustment results of the attention mechanism; the unmanned vehicle drives along the planned path and waits for the drone to charge at the charging intersection; If the drone's energy is consumed too quickly, it will prioritize charging at the nearest charging junction; If the task priority changes, the UAV dynamically adjusts its path to give priority to visiting the new high-priority cluster head; During driving, the unmanned vehicle adjusts its driving speed according to the real-time status of the drone to ensure that it meets the drone at the charging intersection; The task completion and path optimization feedback module is used to enable the drones and unmanned vehicles to return to the base station according to the planned path after the drones complete the visit tasks of all cluster heads, and to feedback the optimization suggestions of the path planning based on the task execution status.

[0008] In a third aspect, the present invention provides a computer device, comprising: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors execute the steps of the method for collaborative path planning between a drone and an unmanned vehicle based on an attention mechanism.

[0009] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for collaborative path planning between a drone and an unmanned vehicle based on an attention mechanism.

[0010] In a fifth aspect, the present invention provides a computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the method for collaborative path planning between a drone and an unmanned vehicle based on an attention mechanism.

[0011] Compared with the prior art, the present invention has the following significant advantages: (1) Low computational complexity; Traditional path planning algorithms (such as genetic algorithms and simulated annealing) have high computational complexity when faced with large-scale, complexly constrained path planning problems. This paper uses a collaborative path planning algorithm based on an attention mechanism to effectively reduce computational complexity, especially when handling large-scale tasks, significantly reducing computing time and resource consumption.

[0012] (2) Strong dynamic adjustment capability; Existing technologies lack dynamic adjustment capabilities, making it difficult to cope with changes in task priorities or rapid energy consumption. This invention, by introducing an attention mechanism, can dynamically adjust the drone's flight path in real time, prioritizing high-priority tasks or paths with lower energy consumption, ensuring efficient task completion.

[0013] (3) Excellent collaborative ability; Traditional collaborative path planning methods for drones and autonomous vehicles often have poor coordination capabilities and struggle to effectively address charging intersections between drones and autonomous vehicles. This invention ensures that drones and autonomous vehicles can charge at the appropriate time and location through precise path planning and charging intersection allocation, significantly improving the efficiency of collaborative operations.

[0014] (4) Path optimization under energy constraints; The present invention fully considers the energy consumption problem of the UAV. Through reasonable path planning and charging strategies, it ensures that the UAV can meet the unmanned vehicle in time to recharge before the energy is exhausted, avoiding mission interruption or failure due to insufficient energy.

[0015] (5) Adapting to complex environments; The algorithm of the present invention can adapt to complex environments and task requirements, especially in large-scale, multi-constrained scenarios, and can still provide an efficient path planning solution with strong robustness and adaptability.

[0016] (6) Minimize task completion time; By optimizing the flight path of the UAV and the driving path of the unmanned vehicle, the present invention can significantly reduce the task completion time, ensure that the data collection task is completed in the shortest time, and improve the overall operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a specific scenario diagram of the method of the present invention.

[0018] Figure 2 Flowchart of the method of the present invention. DETAILED DESCRIPTION

[0019] The present invention is further illustrated below with reference to specific examples. It should be understood that these examples are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, modifications of various equivalent forms of the present invention made by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0020] Example 1 This embodiment provides a collaborative path planning method for drones and unmanned vehicles based on an attention mechanism.

[0021] Specific scenarios of the method in this embodiment: Figure 1 As shown in the figure, an unmanned vehicle (UGV) with unlimited fuel departs from a base station, travels at a constant speed along a planned route, and can also charge unmanned aerial vehicles (UAVs). Simultaneously, N drones depart from the base station and visit the cluster head to collect information. The drones fly at a constant speed, and their energy consumption is proportional to their flight time. Planning a reasonable flight path for the drones, allowing them to rendezvous with the UAVs at appropriate locations to recharge, minimizes the time it takes for the drones to collect information.

[0022] like Figure 2 As shown, the specific implementation steps of the method in this embodiment are as follows: Step 1: Target area and drone flight altitude restrictions; Determine the three-dimensional range of the target area Ω and set the flight altitude limit of the drone to ensure that the drone does not exceed the preset safe altitude range during flight.

[0023] Step 2: sensor deployment and cluster head clustering; In the target area Ω, sensors are deployed to collect environmental information such as sound, temperature, humidity, etc. of the target area Ω, as well as cluster heads for receiving and storing the information data collected by the sensors; It should be noted that enough sensors are deployed in the target area to monitor and obtain information data and transmit the data to the nearby cluster head. The cluster head, with a built-in positioning module, is responsible for receiving and temporarily storing the monitoring data from the sensors. The drone accesses the cluster head to collect information and perform the drone's collection task.

[0024] Step 3: Set the scene parameters: The uniform speed of the unmanned vehicle V_{G}; The charging rate of the unmanned vehicle C_{rate}; The uniform flight speed of the UAV is V_{A}; The energy consumption rate of the drone E_{rate} (energy consumption is proportional to the flight time of the drone); The maximum communication distance D between the UAV and the unmanned vehicle; The maximum communication distance D_{cluster} between the UAV and the cluster head; The scene parameters set in this step are used for the subsequent calculation of the minimum flight time or minimum flight distance of the objective function.

[0025] Step 4: Unmanned vehicle route planning; The unmanned vehicle plans the shortest driving path based on the location of the cluster head. The specific steps are as follows: (1) Perform k-means clustering on the cluster head positions, and the obtained cluster centers are used as the waypoints that the unmanned vehicle needs to visit; (2) Cluster the cluster head positions again to obtain a set of sparse waypoints; (3) With waypoints as vertices, construct and solve a traveling salesman problem (TSP) to find the shortest path for the autonomous vehicle.

[0026] Specifically, the constructed traveling salesman problem TSP is: Undirected graph G'=(V',E'), where V'={0,1,...,k} is the entire vertex set, 0 is the starting vertex; E' is the edge set, representing the connection relationship between vertices; c_{ij} is the non-negative arc cost between vertices i and j; x_{ij} is a binary variable. If the autonomous vehicle travels from i to j, then x_{ij}=1, otherwise it is 0. The constraints of the traveling salesman problem TSP are: The goal is to minimize the total distance traveled by the autonomous vehicle; Ensure that the autonomous vehicle visits each vertex once; Ensure that the unmanned vehicle starts from the base station vertex and ends at the base station vertex; Make sure there is no sub-tour; Constraints on binary decision variables x_{ij}.

[0027] Use greedy algorithm to solve: ① Initial setup: Start from the starting vertex (base station vertex 0) and mark it as visited; ② Select the next vertex: Among the unvisited vertices, select the vertex with the smallest arc cost (distance) to the current vertex as the next vertex to visit; ③Update status: mark the selected vertex as visited and update the current vertex to that vertex; ④ Repeat selection: Repeat steps ② and ③ until all vertices have been visited once; ⑤ Return to the starting vertex: Finally, return to the starting vertex 0 from the last visited vertex to form a complete path.

[0028] Step 5: Collaborative path planning between the unmanned vehicle and the drone; After the unmanned vehicle path planning is completed, a certain number of drones are dispatched to visit the cluster head. Because drones are energy-constrained and need to recharge at specific locations alongside the unmanned vehicles, collaborative path planning for drones and unmanned vehicles is necessary.

[0029] The specific steps are as follows: (1) Allocate a sufficient number of vertices on the path of the unmanned vehicle as the rendezvous and charging locations for the UAV and the unmanned vehicle; (2) Assuming that there are k drones and unmanned vehicles starting from the same base station, a vehicle routing problem (VRP) with energy constraints is constructed, taking into account the energy limit of the drones, the time window allowed for meeting, and the constraints on the drones visiting the vertices of the unmanned vehicle path.

[0030] The constructed vehicle routing problem VRP is: The set of all unmanned vehicle waypoints A={1,2,...,m}; Pre-designated charging intersection point M={m+1,...,n}; The set of all vertices V=M∪A={0,1,2,...,m,m+1,...,n}; The set of all edges E = {(i, j) | i, j∈V, i≠j}; The constraints are: Drones and autonomous vehicles must meet at designated charging intersections; The drone’s flight path must meet energy consumption constraints; The flight path of the drone must be coordinated with the path of the unmanned vehicle to ensure timely charging at the charging intersection.

[0031] Step 6: Path optimization based on attention mechanism; Based on collaborative path planning, an attention mechanism is introduced to optimize the flight path of the UAV. The attention mechanism can help the UAV dynamically adjust its path during flight, giving priority to visiting cluster heads with lower energy consumption or higher task priority.

[0032] The specific implementation steps are: Step 6.1, attention weight allocation; Assign an attention weight to each UAV, which is dynamically adjusted based on the importance of the cluster head, the energy status of the UAV, and the task priority; Assume that the attention weight of the drone is α i , the importance weight of the cluster head is w cluster , the energy state weight of the UAV is w energy , the task priority weight is w priority , then the attention weight can be expressed as: α i =w cluster Importance(i)+w energy EnergyState(i)+w priority Priority(i) Among them: Importance(i) represents the importance of cluster head i; EnergyState(i) represents the current energy state of the UAV; Priority(i) represents the priority of task i.

[0033] Step 6.2, path selection probability calculation; Use the attention mechanism to calculate the path selection probability of the drone at each time step, giving priority to paths with lower energy consumption or higher task priority; Assuming that the probability of the drone choosing path j at time step t is P_{ij}(t), it can be expressed as: in: α i is the attention weight of drone i; α k is the attention weight of drone k; Utility ij represents the utility value of choosing path j at time step t, which can be a function of energy consumption or task priority; Candidates represents the set of paths available for selection at the current time step.

[0034] Step 6.3, dynamically adjust the flight path of the UAV to ensure that it can meet the unmanned vehicle at the charging intersection before the energy is exhausted; The constraint that the drone needs to reach the charging point before its energy is exhausted can be expressed as: E i (t)-E rate ·D charge ≥0 Among them, E i (t) is the current energy of the UAV, E rate is the energy consumption rate, D charge is the distance to the charging intersection.

[0035] Step 7: Path execution and dynamic adjustment; During the actual execution process, drones and unmanned vehicles move according to the planned paths; drones monitor energy consumption in real time during flight and select the optimal path for flight based on the dynamic adjustment results of the attention mechanism; unmanned vehicles travel along the planned paths and wait for drones to charge at the charging intersection.

[0036] Specific dynamic adjustment strategies: Step 7.1, energy consumption monitoring and charging priority adjustment; If the drone's energy is consumed too quickly, it will prioritize charging at the nearest charging junction; Assume that the current energy of the drone is E i (t), energy consumption rate is E rate , the dynamic monitoring of energy consumption can be expressed as: E i (t+1)=E i (t)-E rate ·∆t Where ∆t is the time step, if E i ( t +1) <Threshold , then the nearest charging intersection j* is preferred: Step 7.2: Task priority changes and path adjustments; If the task priority changes, the UAV dynamically adjusts its path to give priority to visiting the new high-priority cluster head; Assuming that the task priority is updated to Priority'(t), the path adjustment can be expressed as: Step 7.3: Adjust the speed of the unmanned vehicle; During driving, the unmanned vehicle adjusts its driving speed according to the real-time status of the drone to ensure that it meets the drone at the charging intersection; Assume that the current speed of the autonomous vehicle is V G , the adjusted speed is V G ’ , then the speed adjustment can be expressed as: Where ∆V is the velocity increment that is dynamically adjusted according to the real-time status of the UAV (such as energy status or mission progress).

[0037] Step 8: Task completion and path optimization feedback; When the UAV completes the visit mission to all cluster heads, the UAV and the unmanned vehicle return to the base station according to the planned path; based on the task execution status, the optimization suggestions for the path planning are fed back.

[0038] Feedback mechanism: Step 8.1, path deviation record; Record the deviation between the actual path of the UAV and the planned path. Assume that the actual path of the UAV is Path actual , the planned path is Path planned , then the path deviation can be expressed as: Here, i represents a node or time step on the path.

[0039] Step 8.2: Analyze the causes of deviations; Analyze the causes of deviation, such as excessive energy consumption, changes in task priority, etc. Assume that the causes of deviation can be classified as energy deviation ΔE and task priority change Δ P , it can be expressed as: Where: E initial is the initial energy of the drone; E actual (t) is the actual remaining energy of the UAV at time t; Priority planned (i) is the planning task priority; Priority actual (i) is the actual task priority.

[0040] Step 8.3, attention weight adjustment; According to the feedback results, the weight distribution strategy of the attention mechanism is adjusted to optimize the path planning algorithm. Assuming that the adjustment amount of the attention weight is Δα, it can be expressed as: in:η is the learning rate, which is used to control the magnitude of weight adjustment; w energy is the weight coefficient of energy deviation; w priority is the weight coefficient of task priority change; The adjusted attention weight is: α i '=α i +∆α.

[0041] Step 8.4: Optimize the path planning algorithm; Optimize the path planning algorithm based on the feedback results to ensure further improvement of the efficiency and adaptability of path planning in future tasks. Assume that the optimized path planning objective function is Objective new , it can be expressed as: Where Deviation(i) is the path deviation of task i; CompletionTime(i) is the completion time of task i; λ is the weight coefficient that balances path deviation and task completion time.

[0042] Among them, the feedback result is a comparative analysis of actual execution data and planning data, which is used to dynamically adjust algorithm parameters (such as attention weight) and optimize subsequent path planning to improve collaborative efficiency and task adaptability.

[0043] The above attention-based collaborative path planning algorithm for drones and unmanned vehicles can achieve efficient path planning in complex environments, ensuring that drones can complete tasks under energy constraints and work in collaboration with unmanned vehicles to reduce task completion time.

[0044] Example 2 Based on the same inventive concept, the present application also provides a drone and unmanned vehicle collaborative path planning system based on an attention mechanism. The implementation solution provided by this drone and unmanned vehicle collaborative path planning system based on an attention mechanism is similar to the implementation solution described in the method of Example 1. Therefore, the specific limitations of one or more of the following drone and unmanned vehicle collaborative path planning system based on an attention mechanism can be referred to the limitations of the method in Example 1 and will not be repeated here.

[0045] In an exemplary embodiment, a collaborative path planning system for a UAV and an unmanned vehicle based on an attention mechanism is provided, comprising: The target area determination module is used to determine the three-dimensional range of the target area Ω and set the flight altitude limit of the UAV.

[0046] Sensors are deployed in the target area Ω to collect information data of the target area Ω.

[0047] The cluster head is deployed in the target area Ω and is used to receive and store the information data collected by the sensors.

[0048] Parameter setting module, used to set scene parameters: The uniform driving speed of the unmanned vehicle V_{G}; the charging rate of the unmanned vehicle C_{rate}; the uniform flight speed of the drone V_{A}; the energy consumption rate of the drone E_{rate}; the maximum communication distance D between the drone and the unmanned vehicle; the maximum communication distance D_{cluster} between the drone and the cluster head.

[0049] Unmanned vehicle route planning module, used for: Perform k-means clustering on the cluster head positions, and the obtained cluster centers are used as the waypoints that the unmanned vehicle needs to visit; Cluster the cluster head positions again to obtain a set of sparse waypoints; Using waypoints as vertices, construct and solve a traveling salesman problem (TSP) to find the shortest path for the autonomous vehicle. Among them, the constructed traveling salesman problem TSP is: Undirected graph G'=(V',E'), where V'={0,1,...,k} is the entire vertex set, 0 is the starting vertex; E' is the edge set, representing the connection relationship between vertices; c_{ij} is the non-negative arc cost between vertices i and j; x_{ij} is a binary variable. If the autonomous vehicle travels from i to j, then x_{ij}=1, otherwise it is 0. Among them, the constraints of the traveling salesman problem TSP are: The goal is to minimize the total distance traveled by the autonomous vehicle; Ensure that the autonomous vehicle visits each vertex once; Ensure that the unmanned vehicle starts from the base station vertex and ends at the base station vertex; Make sure there is no sub-tour; Constraints on binary decision variables x_{ij}.

[0050] Unmanned vehicle and drone collaborative path planning module for: Allocate a sufficient number of vertices on the path of the unmanned vehicle as the rendezvous and charging locations for the UAV and the unmanned vehicle; Assuming that there are k drones and unmanned vehicles starting from the same base station, a vehicle routing problem (VRP) with energy constraints is constructed, taking into account the energy limit of the drones, the time window allowed for meeting, and the constraints on the drones visiting the vertices of the unmanned vehicle's path; Among them, the constructed vehicle routing problem VRP is: The set of all unmanned vehicle waypoints A={1,2,...,m}; Pre-designated charging intersection point M={m+1,...,n}; The set of all vertices V=M∪A={0,1,2,...,m,m+1,...,n}; The set of all edges E = {(i, j) | i, j∈V, i≠j}; The constraints are: Drones and autonomous vehicles must meet at designated charging intersections; The drone’s flight path must meet energy consumption constraints; The flight path of the drone must be coordinated with the path of the unmanned vehicle to ensure timely charging at the charging intersection.

[0051] Attention mechanism optimization module, used for: Based on the collaborative path planning of unmanned vehicles and drones, an attention mechanism is introduced to optimize the flight path of drones. Among them, the implementation method of the attention mechanism optimization module is: Assign an attention weight to each UAV, which is dynamically adjusted based on the importance of the cluster head, the energy status of the UAV, and the task priority; Use the attention mechanism to calculate the path selection probability of the drone at each time step, giving priority to paths with lower energy consumption or higher task priority; Dynamically adjust the drone's flight path to ensure it can meet the unmanned vehicle at the charging intersection before its energy is exhausted.

[0052] Path execution and dynamic adjustment module, used to: During the actual execution process, the drone and the unmanned vehicle move along the planned path. The drone monitors energy consumption in real time during flight and selects the optimal path based on the dynamic adjustment results of the attention mechanism. The unmanned vehicle drives along the planned path and waits for the drone to charge at the charging intersection. If the drone's energy is consumed too quickly, it will prioritize charging at the nearest charging junction; If the task priority changes, the UAV dynamically adjusts its path to give priority to visiting the new high-priority cluster head; While driving, the unmanned vehicle adjusts its driving speed according to the real-time status of the drone to ensure that it meets the drone at the charging intersection.

[0053] Task completion and path optimization feedback module, used to: After the UAV completes the visit mission to all cluster heads, it will make the UAV and the unmanned vehicle return to the base station according to the planned path, and feedback the optimization suggestions of the path planning based on the task execution status; In the task completion and path optimization feedback module, the optimization suggestions for feedback path planning include: The deviation between the actual path of the UAV and the planned path is recorded, and the cause of the deviation is analyzed as feedback. Based on the feedback results, the weight distribution strategy of the attention mechanism is adjusted to optimize the path planning algorithm.

[0054] Example 3 This embodiment provides a method for graphing system logs: when performing cloud computing system operation and maintenance, industrial Internet of Things system fault diagnosis, and financial system security monitoring, the attention mechanism-based drone and unmanned vehicle collaborative path planning method described in Example 1 is used to graph the logs of the corresponding systems.

[0055] Example 4 Each module in the above system can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0056] In an exemplary embodiment, a computer device is provided, which may be a terminal. The computer device further includes a processor, memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and computer program stored in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via wired or wireless communication, which may be achieved via Wi-Fi, a mobile cellular network, NFC (near-field communication), or other technologies. When executed by the processor, the computer program implements the steps of a method for collaborative path planning between a drone and an unmanned vehicle based on an attention mechanism. The display unit of the computer device is used to produce a visual image and may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0057] Those skilled in the art will understand that the structure of the above-mentioned computer device is only a partial structure related to the solution of the present application and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components, or have a different component arrangement.

[0058] In an exemplary embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for collaborative path planning between a drone and an unmanned vehicle based on an attention mechanism are implemented.

[0059] In an exemplary embodiment, a computer program product is also provided, including a computer program / instruction, characterized in that when the computer program / instruction is executed by a processor, the steps of a method for collaborative path planning between a drone and an unmanned vehicle based on an attention mechanism are implemented.

[0060] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processors (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), data processing logic devices based on quantum computing, and the like.

[0061] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0062] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A collaborative path planning method for UAVs and unmanned vehicles based on attention mechanism, characterized in that: The steps include: Step 1: Determine the three-dimensional range of the target area Ω and set the flight altitude limit of the UAV; Step 2: Deploy sensors for collecting information data of the target area Ω and cluster heads for receiving and storing the information data collected by the sensors within the target area Ω. Step 3: Set the scene parameters: The uniform speed of the unmanned vehicle V_{G}; The charging rate of the unmanned vehicle C_{rate}; The uniform flight speed of the UAV is V_{A}; The energy consumption rate of the drone E_{rate}; The maximum communication distance D between the UAV and the unmanned vehicle; The maximum communication distance D_{cluster} between the UAV and the cluster head; Step 4: Unmanned vehicle route planning; Perform k-means clustering on the cluster head positions, and the obtained cluster centers are used as the waypoints that the unmanned vehicle needs to visit; Using waypoints as vertices, construct and solve a traveling salesman problem (TSP) to find the shortest path for the autonomous vehicle. The constructed traveling salesman problem TSP is: Undirected graph G'=(V',E'), where V'={0,1,...,k} is the entire vertex set and 0 is the starting vertex; E' is the edge set, which represents the connection relationship between vertices; c_{ij} is the non-negative arc cost between vertices i and j; x_{ij} is a binary variable. If the autonomous vehicle travels from i to j, then x_{ij}=1, otherwise it is 0. The constraints of the traveling salesman problem TSP are: The goal is to minimize the total distance traveled by the autonomous vehicle; Ensure that the autonomous vehicle visits each vertex once; Ensure that the unmanned vehicle starts from the base station vertex and ends at the base station vertex; Make sure there is no sub-tour; Constraints on binary decision variables x_{ij}; Step 5: Collaborative path planning between the unmanned vehicle and the drone; Allocate a sufficient number of vertices on the path of the unmanned vehicle as the rendezvous and charging locations for the UAV and the unmanned vehicle; Assuming that there are k drones and unmanned vehicles starting from the same base station, a vehicle routing problem (VRP) with energy constraints is constructed, taking into account the energy limit of the drones, the time window allowed for meeting, and the constraints on the drones visiting the vertices of the unmanned vehicle's path; The constructed vehicle routing problem VRP is: The set of all unmanned vehicle waypoints A={1,2,...,m}; Pre-designated charging intersection point M={m+1,...,n}; The set of all vertices V=M∪A={0,1,2,...,m,m+1,...,n}; The set of all edges E = {(i, j) | i, j∈V, i≠j}; The constraints are: Drones and autonomous vehicles must meet at designated charging intersections; The drone’s flight path must meet energy consumption constraints; The drone’s flight path must be coordinated with the path of the unmanned vehicle to ensure timely charging at the charging intersection; Step 6: Based on the collaborative path planning of the unmanned vehicle and the drone, an attention mechanism is introduced to optimize the flight path of the drone. In step 6.1, an attention weight is assigned to each UAV, and the weight is dynamically adjusted according to the importance of the cluster head, the energy status of the UAV, and the task priority; In step 6.2, the attention mechanism is used to calculate the path selection probability of the UAV at each time step, giving priority to paths with lower energy consumption or higher task priority; Step 6.3, dynamically adjust the flight path of the UAV to ensure that it can meet the unmanned vehicle at the charging intersection before the energy is exhausted; Step 7: During the actual execution, the drone and the unmanned vehicle move along the planned path. The drone monitors energy consumption in real time during flight and selects the optimal path based on the dynamic adjustment results of the attention mechanism. The unmanned vehicle follows the planned path and waits for the drone to charge at the charging intersection. If the drone's energy is consumed too quickly, it will prioritize charging at the nearest charging junction; If the task priority changes, the UAV dynamically adjusts its path to give priority to visiting the new high-priority cluster head; During driving, the unmanned vehicle adjusts its speed according to the real-time status of the drone to ensure that it meets the drone at the charging intersection; Step 8: After the UAV completes the visit mission to all cluster heads, the UAV and the unmanned vehicle return to the base station according to the planned path; based on the task execution status, the optimization suggestions for the path planning are fed back.

2. The method for collaborative path planning between a UAV and an unmanned vehicle based on an attention mechanism according to claim 1, characterized in that: In step 8, the feedback path planning optimization suggestions include: The deviation between the actual path of the UAV and the planned path is recorded, and the cause of the deviation is analyzed as feedback. Based on the feedback results, the weight distribution strategy of the attention mechanism is adjusted to optimize the path planning algorithm.

3. A collaborative path planning system for drones and unmanned vehicles based on an attention mechanism, characterized by: include: The target area determination module is used to determine the three-dimensional range of the target area Ω and set the flight altitude limit of the UAV; Sensors are deployed in the target area Ω to collect information data of the target area Ω; Cluster head, deployed in the target area Ω, is used to receive and store information data collected by sensors; The parameter setting module is used to set the scenario parameters: the uniform driving speed of the unmanned vehicle V_{G}; the charging rate of the unmanned vehicle C_{rate}; the uniform flight speed of the drone V_{A}; the energy consumption rate of the drone E_{rate}; the maximum communication distance D between the drone and the unmanned vehicle; the maximum communication distance D_{cluster} between the drone and the cluster head; Unmanned vehicle route planning module, used for: Perform k-means clustering on the cluster head positions, and the obtained cluster centers are used as the waypoints that the unmanned vehicle needs to visit; Cluster the cluster head positions again to obtain a set of sparse waypoints; Using waypoints as vertices, construct and solve a traveling salesman problem (TSP) to find the shortest path for the autonomous vehicle. Among them, the constructed traveling salesman problem TSP is: Undirected graph G'=(V',E'), where V'={0,1,...,k} is the entire vertex set, 0 is the starting vertex; E' is the edge set, representing the connection relationship between vertices; c_{ij} is the non-negative arc cost between vertices i and j; x_{ij} is a binary variable. If the autonomous vehicle travels from i to j, then x_{ij}=1, otherwise it is 0. Among them, the constraints of the traveling salesman problem TSP are: The goal is to minimize the total distance traveled by the autonomous vehicle; Ensure that the autonomous vehicle visits each vertex once; Ensure that the unmanned vehicle starts from the base station vertex and ends at the base station vertex; Make sure there is no sub-tour; Constraints on binary decision variables x_{ij}; Unmanned vehicle and drone collaborative path planning module for: Allocate a sufficient number of vertices on the path of the unmanned vehicle as the rendezvous and charging locations for the UAV and the unmanned vehicle; Assuming that there are k drones and unmanned vehicles starting from the same base station, a vehicle routing problem (VRP) with energy constraints is constructed, taking into account the energy limit of the drones, the time window allowed for meeting, and the constraints on the drones visiting the vertices of the unmanned vehicle's path; Among them, the constructed vehicle routing problem VRP is: The set of all unmanned vehicle waypoints A={1,2,...,m}; Pre-designated charging intersection point M={m+1,...,n}; The set of all vertices V=M∪A={0,1,2,...,m,m+1,...,n}; The set of all edges E = {(i, j) | i, j∈V, i≠j}; The constraints are: Drones and autonomous vehicles must meet at designated charging intersections; The drone’s flight path must meet energy consumption constraints; The drone’s flight path must be coordinated with the path of the unmanned vehicle to ensure timely charging at the charging intersection; The attention mechanism optimization module is used to introduce the attention mechanism to optimize the flight path of the drone based on the collaborative path planning of the unmanned vehicle and the drone; Among them, the implementation method of the attention mechanism optimization module is: Assign an attention weight to each UAV, which is dynamically adjusted based on the importance of the cluster head, the energy status of the UAV, and the task priority; Use the attention mechanism to calculate the path selection probability of the drone at each time step, giving priority to paths with lower energy consumption or higher task priority; Dynamically adjust the drone's flight path to ensure it can meet the unmanned vehicle at the charging intersection before its energy is exhausted; The path execution and dynamic adjustment module is used to: During the actual execution process, the drone and the unmanned vehicle move according to the planned path; the drone monitors energy consumption in real time during flight and selects the optimal path based on the dynamic adjustment results of the attention mechanism; the unmanned vehicle drives along the planned path and waits for the drone to charge at the charging intersection; If the drone's energy is consumed too quickly, it will prioritize charging at the nearest charging junction; If the task priority changes, the UAV dynamically adjusts its path to give priority to visiting the new high-priority cluster head; During driving, the unmanned vehicle adjusts its speed according to the real-time status of the drone to ensure that it meets the drone at the charging intersection; The task completion and path optimization feedback module is used to enable the drones and unmanned vehicles to return to the base station according to the planned path after the drones complete the visit tasks of all cluster heads, and to feedback the optimization suggestions of the path planning based on the task execution status.

4. The UAV and UGV collaborative path planning system based on attention mechanism according to claim 3 is characterized in that: In the task completion and path optimization feedback module, the optimization suggestions for feedback path planning include: The deviation between the actual path of the UAV and the planned path is recorded, and the cause of the deviation is analyzed as feedback. Based on the feedback results, the weight distribution strategy of the attention mechanism is adjusted to optimize the path planning algorithm.

5. A computer device, characterized in that: include: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors execute the steps of the method for collaborative path planning between a drone and an unmanned vehicle based on an attention mechanism as described in any one of claims 1-2.

6. A computer-readable storage medium storing a computer program, characterized in that: When the program is executed by a processor, the steps of the method for collaborative path planning between a drone and an unmanned vehicle based on an attention mechanism are implemented as described in any one of claims 1-2.

7. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by the processor, the steps of the method for collaborative path planning between a drone and an unmanned vehicle based on an attention mechanism are implemented as described in any one of claims 1-2.

Images