A collaborative road network search method for heterogeneous UAVs based on Euler paths

Abstract

The present invention discloses a heterogeneous UAV collaborative road network search method based on Euler path, comprising the following steps: step 1, constructing an undirected graph: constructing a city road network as a weighted undirected graph according to the road structure information of the city; step 2, determining the Euler graph; step 3, constructing the Euler graph: adding duplicate edges to the vertices with odd degrees, changing the degrees to even numbers, and completing the construction of the Euler graph; step 4, searching for Euler paths: finding Euler paths in the Euler graph using depth-first search; step 5, dividing and allocating Euler paths; step 6, selecting a starting point: selecting a starting point of a search section for the UAV, and then starting to perform a search task. The present invention improves the effectiveness and safety of UAV search, and at the same time fully utilizes the characteristics of the road network to improve the search efficiency.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicle collaborative search path planning, and in particular relates to a heterogeneous unmanned aerial vehicle collaborative road network search method based on Euler paths. Background Art

[0002] With the development and maturity of drone technology, drones have played an increasingly important role in many fields, such as logistics distribution, environmental monitoring, security monitoring and disaster response. Especially in the search tasks of urban road networks and transportation networks, drones have become an attractive solution with their fast, flexible and efficient characteristics. However, in complex search tasks, existing path planning technologies face many challenges, especially the difficulties brought by the heterogeneity of drones and the particularity of search tasks, which may lead to low efficiency of task execution and waste of resources. In order to solve the planning problem in drone road network search, Chinese patent 117889882A discloses a road network planning method based on ground-air unmanned cluster collaborative situation assessment, which does not make full use of the characteristics of the road network, so it is difficult to ensure full coverage of the search area; Chinese patent 118274846B records a drone group collaborative path planning method and system for road hidden danger inspection, which simulates the biological evolution process and iteratively optimizes the path, trying to shorten the total path length and improve the search efficiency. This method has achieved certain results in homogeneous drone systems, but has limitations in heterogeneous drone scenarios. In addition, for example, Chinese patent document 109520504B records a method for optimizing the inspection path of unmanned aerial vehicles based on grid discretization. It adopts a partitioned task allocation strategy to improve the coordination ability of multi-unmanned aerial vehicle systems to a certain extent, but does not fully consider the heterogeneous characteristics, and resource allocation and path coverage are still not efficient enough. Although the above methods have alleviated the problem of unmanned aerial vehicle path planning to a certain extent, there are still many shortcomings. First, the method based on genetic algorithm takes a long time to solve complex search problems, and it is difficult to ensure the global optimality of the path within polynomial time. Secondly, the existing methods often ignore the coordination ability of heterogeneous unmanned aerial vehicle systems, fail to fully utilize the differences in speed, endurance time and load capacity, resulting in low resource utilization. Finally, although some methods take into account the characteristics of the road network, they still have deficiencies in full coverage of search tasks and optimization of the total length of the path, and it is difficult to meet the needs of efficient task completion. Therefore, how to achieve efficient and comprehensive road network search in heterogeneous unmanned aerial vehicle systems is still a technical problem that needs to be solved urgently. Summary of the invention

[0003] The purpose of the present invention is to provide a heterogeneous UAV collaborative road network search method based on Euler path, so as to solve the limitations of the prior art, improve effectiveness, enhance adaptability and reduce costs.

[0004] In order to achieve the purpose of the present invention, the present invention provides a heterogeneous UAV collaborative road network search method based on Euler path, and the method comprises the following steps:

[0005] Step 1: Construct an undirected graph: construct the urban road network into a weighted undirected graph based on the road structure information of the city;

[0006] Step 2, determining whether the Euler graph is an Euler graph: determining whether the undirected graph is an Euler graph according to the degree of the information statistics nodes of the undirected graph;

[0007] Step 3, constructing an Euler graph: when it is determined that the undirected graph is not the Euler graph, adding duplicate edges to the vertices with odd degrees to change the degrees to even numbers, thereby completing the construction of the Euler graph;

[0008] Step 4, search for the Euler path: use depth-first search to find the Euler path in the Euler graph, ensuring full coverage of the road network and the shortest total distance;

[0009] Step 5, dividing and allocating the Euler path: dividing and allocating the Euler path according to the speed ratio relationship;

[0010] Step 6: Select the starting point: Select the task starting point in the assigned section for each drone by calculating the endpoint closer to the current position, use any path planning algorithm to navigate to the starting point, and start the search task.

[0011] Compared with the prior art, the significant improvements of the present invention are: (1) Comprehensive coverage: The road network search method of the present invention can make full use of the characteristics of the road network and apply the theory of Euler path to ensure comprehensive coverage of the search area and the shortest total distance; (2) Efficiency improvement: By considering the heterogeneity of drones, the present invention can effectively plan collaborative search paths and make full use of faster drones, thereby avoiding the problems of insufficient resource utilization and low search efficiency and improving search efficiency; (3) Fast response: The algorithm used in the present invention is different from the genetic algorithm commonly used in similar technologies. It does not require iterative optimization and can simply and efficiently perform searches within a limited time complexity, quickly respond to the needs of search tasks, and improve the speed of task execution.

[0012] In order to more clearly illustrate the functional characteristics and structural parameters of the present invention, further description is given below in conjunction with the accompanying drawings and specific implementation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0014] Figure 1 It is a technical solution flow chart of the heterogeneous UAV collaborative road network search of the present invention;

[0015] Figure 2 It is a schematic diagram of the result of adding repeated edges using the Edmonds-Johnson algorithm in step 3 of the present invention;

[0016] Figure 3 is a result schematic diagram of the Euler path found by using the Hierholzer algorithm in step 4 of the present invention;

[0017] Figure 4 (a), (b), and (c) are steps 5 of the present invention to divide the Euler path into , , Schematic diagram of the results of three segments, which are assigned to drones numbered 1, 2, and 3 respectively. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0019] The present invention provides a heterogeneous UAV collaborative road network search method based on Euler path, combined with Figure 1 , including the following steps:

[0020] Step 1: Construct an undirected graph: construct the urban road network into a weighted undirected graph based on the road structure information of the city;

[0021] The weighted undirected graph is:

[0022] ;

[0023] in, is a set of nodes, representing intersections, is a set of edges, representing streets, is a weight function, where the weight represents the length of the street, The edges can be mapped to real numbers representing their lengths, namely .

[0024] Step 2, determining whether the Euler graph is an Euler graph: according to the information statistics of the undirected graph, the degree of the nodes, that is, the number of edges connected to each node, is determined to determine whether the undirected graph is an Euler graph;

[0025] The determining of the undirected graph When determining whether it is the Euler graph, if the number of nodes with odd degrees is greater than 2, it is not the Euler graph.

[0026] Step 3, constructing an Euler graph: when it is determined that the undirected graph is not the Euler graph, adding duplicate edges to the vertices with odd degrees to change the degrees to even numbers, thereby completing the construction of the Euler graph;

[0027] The Edmonds-Johnson algorithm is used to minimize the total weight of the duplicate edges, and the duplicate edges are added to the undirected graph. The following are the specific steps of the Edmonds-Johnson algorithm in this patent:

[0028] 3-1. Identify odd-degree vertices and collect the vertices with odd degrees. These vertices are key points that need to be processed in subsequent steps.

[0029] 3-2. Construct a complete graph: Based on the odd number of vertices, construct a new complete graph, in which each vertex is connected to all other vertices, and the edge weights of the complete graph are calculated using the Dijkstra algorithm in step 3-3, that is, in the undirected graph Find the shortest path length between any two odd-degree vertices;

[0030] 3-3. Run Dijkstra algorithm: For each pair of vertices in the complete graph, take any one of them as the starting point and the other as the end point, and combine the undirected graph The information in the graph is used to calculate the shortest path length between two points using the Dijkstra algorithm. The Dijkstra algorithm is a greedy algorithm that gradually finds the shortest path to all vertices in the graph by maintaining a priority queue.

[0031] 3-4. Finding the minimum weight perfect matching: In the complete graph, use the minimum weight perfect matching algorithm to find a set of edges that connects each pair of vertices, and no two edges share a vertex. At the same time, the total weight of this set of edges is the smallest. This matching process can be implemented by a variety of algorithms, such as the Hungarian algorithm or the auction algorithm.

[0032] 3-5. Determine the duplicate edges that need to be added: According to the minimum weight perfect matching, determine which edges are the undirected graph The repeated edges that need to be added in the undirected graph will connect the The odd-degree vertices in make the degrees of all vertices become even.

[0033] 3-6: Add duplicate edges to the undirected graph : Add the duplicate edge to the undirected graph , update the undirected graph The edge set and vertex degrees of the vertices ensure that the total length of the added edges is minimal;

[0034] 3-7: Verify the undirected graph Euler property: After adding the repeated edges, recheck the degree of each vertex of the graph to ensure that the degrees of all vertices are even; if the degrees of all vertices are even, the graph has become an Euler graph, and a closed path that passes through each edge at least once can be found, namely the Euler path.

[0035] Combination Figure 2 , the result after adding repeated edges; the algorithm can ensure that in the process of converting a non-Eulerian graph to an Euler graph, the total length of the added repeated edges is minimized, thus providing an optimization basis for the subsequent Euler path search.

[0036] Step 4, search for an Euler path: use depth-first search to find an Euler path in the Euler graph that passes through each edge exactly once, ensuring full coverage of the road network and the shortest total distance;

[0037] The Hierholzer algorithm is used to plan the Euler path in the Euler graph. The algorithm is implemented based on depth-first search. The key is to ensure that a loop that passes through each edge exactly once can be found in the Euler graph, so it can ensure full coverage of the road network and the shortest total distance, and only requires linear time complexity:

[0038] ;

[0039] in, is the number of edges in the Euler graph;

[0040] 4-1: From the undirected graph that is already an Euler graph Choose any vertex as the starting point.

[0041] 4-2: Starting from the selected starting point, perform a depth-first search along the edges to visit adjacent vertices. During the search, each time an edge is passed, it is deleted from the graph.

[0042] 4-3: If a vertex is reached during the depth-first search and all edges of the vertex have been visited (that is, the degree of the vertex is 0), then the vertex is pushed onto the stack.

[0043] 4-4: When the current vertex has no unvisited edges, pop a vertex from the stack and continue searching. This process is actually backtracking, looking for a path back to the starting point.

[0044] 4-5: Repeat steps 4-2 to 4-4 until the stack is empty. At this point, the vertex sequence in the stack is output in reverse order, which is the Euler path starting from the starting point.

[0045] 4-6: Check the undirected graph Are there any other unvisited edges in the Eulerian path? If so, start from any endpoint of these edges and repeat the above process to merge the new loop found into the existing Euler path.

[0046] 4-7: When the undirected graph When all edges in have been visited and the stack is empty, the algorithm ends. At this point, the obtained sequence is a complete Euler path; combined with Figure 3 , the starting point of the path is point 0, and the arrow indicates the direction of the Euler path when it passes through each edge.

[0047] Step 5, dividing and allocating the Euler path: dividing and allocating the Euler path according to the speed ratio relationship;

[0048] 5-1. Represent the Euler path with nodes and calculate the sum of the velocities of the drone;

[0049] Represent the Euler path with nodes :

[0050] ;

[0051] in, is every node on the path;

[0052] Calculate the total weight of the loop :

[0053]

[0054] Calculate the speed of the drone and :

[0055] ;

[0056] in, It is the number of the drone. is the speed of different drones, It is accumulated from the first drone to the The sum of the speeds of the racks;

[0057] 5-2. Find the path division boundary points of the drone as needed;

[0058] For the number The drone finds the node along the Euler path , the node Need to meet:

[0059] and .

[0060] 5-3. Determine the ownership of the boundary points based on the principle of minimizing the total task time;

[0061] when When The path endpoint ,otherwise .

[0062] 5-4, for UAV distribution Segment path.

[0063] ;

[0064] ;

[0065] …;

[0066] ;

[0067] in, is the path assigned by the first drone, is the node on the Euler path Start to Node of road sections, is each node passed on the road segment, ,…, are the road sections assigned to different drones, Passing Node , Passing Node .when When the road segment division and allocation results are combined Figure 4 , Figure 4 (a), (b), and (c) correspond to road sections respectively , , , used to assign to drones numbered 1, 2, and 3.

[0068] Step 6, select the starting point: select the starting point of the search section for each drone, and then start the search task; calculate the endpoint of the section that is closer to the current position in a straight line (such as The endpoint is and ), select the mission starting point in the assigned section for each drone, use any path planning algorithm (such as dijkstra, A*) to navigate from the current position to the starting point, and start the search mission. After the drone reaches the starting point of the search mission, the perception system is started, and various sensors and algorithms can be used to perceive the surrounding environment.

[0069] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0070] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A heterogeneous UAV collaborative road network search method based on Euler path, characterized in that: The following steps are involved: Step 1: Construct an undirected graph: construct the urban road network into a weighted undirected graph based on the road structure information of the city; Step 2, determining whether the Euler graph is an Euler graph: determining whether the undirected graph is an Euler graph according to the degree of the information statistics nodes of the undirected graph; Step 3, constructing an Euler graph: when it is determined that the undirected graph is not the Euler graph, adding duplicate edges to the vertices with odd degrees to change the degrees to even numbers, thereby completing the construction of the Euler graph; Step 4, search for the Euler path: use depth-first search to find the Euler path in the Euler graph, ensuring full coverage of the road network and the shortest total distance; Step 5, dividing and allocating the Euler path: dividing and allocating the Euler path according to the speed ratio relationship; 5-1. Represent the Euler path with nodes and calculate the sum of the velocities of the drone; Specifically: Represent the Euler path with nodes : ； in, is every node on the path; Calculate the total weight of the Euler path : ； in, is the total number of nodes on the Euler path, is the node number, Is a node and nodes The weight of the edge between them; Calculate the speed of the drone and : ； in, is the number of the drone. is the speed of different drones, It is accumulated from the first drone to the The sum of the speeds of the racks; 5-2. Find the path division boundary points of the drone as needed; Specifically: For the number The drone finds the node along the Euler path , the node Need to meet: and ; 5-3. Determine the ownership of the boundary points based on the principle of minimizing the total task time; 5-4, for UAV distribution Segment path; Step 6: Select the starting point: By calculating the endpoint of the road segment closer to the current position, select the task starting point in the assigned road segment for each drone, use any path planning algorithm to navigate to the starting point, and start the search task.

2. According to the method of claim 1, the method is characterized in that: The weighted undirected graph in step 1 is: ； in, is a collection of nodes, is the edge set, is the weight function.

3. The method for searching heterogeneous UAV cooperative road networks based on Euler paths according to claim 2 is characterized in that: The step 2 determines the undirected graph When determining whether it is the Euler graph, if the number of nodes with odd degrees is greater than 2, it is not the Euler graph.

4. The method for searching heterogeneous UAV cooperative road networks based on Euler paths according to claim 1 is characterized in that: The step 3 minimizes the total weight of the repeated edges using the Edmonds-Johnson algorithm and adds the repeated edges to the undirected graph.

5. The method for searching heterogeneous UAV cooperative road networks based on Euler paths according to claim 1 is characterized in that: The step 4 is to plan the Euler path in the Euler graph using the Hierholzer algorithm: ； in, is the number of edges in the Euler graph.

6. The method for searching heterogeneous UAV cooperative road networks based on Euler paths according to claim 1 is characterized in that: Step 5-3: when When The path endpoint ,otherwise .

7. The method for searching heterogeneous UAV cooperative road networks based on Euler paths according to claim 1 is characterized in that: Step 5-4: ； ； …； ； in, is the path assigned by the first drone, is the node on the Euler path Start to Node of road sections, is each node passed on the road segment, ,…, These are the road sections to which different drones are assigned.

Images