A path planning method and system based on a connected search inflection point

Abstract

The application provides a path planning method and system based on a connected search inflection point, and the method comprises the following steps: constructing an obstacle map and a target set, storing a starting point and an ending point as a target point pair in the target set; connecting the target point pair, calculating the directed distance from the vertex of the obstacle intersecting with the connected line to the connected line, and selecting the vertex with the maximum absolute value on the positive and negative sides of the connected line as an effective vertex; if there is no intersecting obstacle, deleting the point pair from the target set and adding it to a result set; calculating the obstacle weight, determining the maximum weight obstacle, calculating the comprehensive cost of the effective inflection point of the obstacle, selecting the inflection point with a smaller cost as the optimal inflection point, and updating the target set; repeating the steps until the target set is empty, connecting the target point pairs in the result set, and obtaining the path planning result. The method has good stability and adaptability, is suitable for different types of maps, greatly reduces the total turning number and total turning angle of the path, and reduces the total distance and calculation time.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of path planning of engineering vehicles, and particularly relates to a path planning method and system based on a connected search inflection point. BACKGROUND

[0002] In the modern engineering field, articulated large engineering vehicles such as loaders, mine trucks and articulated dump trucks have become indispensable work equipment in construction engineering, mining, logistics and warehousing scenes. With their large load capacity, high passability and flexible steering characteristics, they undertake core tasks such as material transportation, site leveling and ore loading in complex terrain. However, as the engineering scale continues to expand and the intelligent transformation accelerates, the path planning algorithms in traditional assisted driving and autonomous driving technology have been unable to meet the actual needs of articulated engineering vehicles in high-density work scenes such as narrow mines, dynamic stockyards and underground tunnels. Frequent steering, circuitous path, and calculation delay problems occur frequently, which not only restricts efficiency and energy economy, but also directly affects the stability and safety of vehicle operation, and targeted technical innovation is urgently needed.

[0003] Traditional algorithms and their improved solutions perform well in general mobile robot path planning, and their core logic is to generate a discrete polyline path on a rasterized map through a node expansion mechanism to achieve obstacle avoidance. In recent years, researchers have also tried to alleviate some problems in the path planning of articulated engineering vehicles through path post-processing methods, such as using spline curve smoothing, steering point clustering and other methods for local optimization, such as Bezier curve smoothing to reduce the number of steering times to a certain extent, and dynamic programming to reduce the cumulative steering angle.

[0004] However, traditional algorithms and the motion characteristics of articulated engineering vehicles are fundamentally incompatible. Because these vehicles use an articulated body structure, the front and rear vehicle bodies need to be controlled simultaneously when steering, resulting in a large steering radius and significant inertia. In addition, the steering angle is limited by the mechanical structure, and the discrete polyline path generated by existing algorithms has problems such as excessive number of steering points, excessive total steering angle, and inability to calculate normally when facing complex irregular terrain. Furthermore, existing path planning solutions lack adaptation to the special nature of engineering vehicle work scenes, such as not considering the swept space when the articulated vehicle body steers in narrow passages on construction sites, which may plan a theoretically feasible but practically impossible path. In the slope transportation scene of open-pit mines, the distribution of steering points is not optimized for slope changes, which can cause power interruption and even slope sliding risks for vehicles. The optimization of path length is insufficient, and there are problems such as redundant travel distance. Although there are path post-processing methods, these methods can only locally optimize the path shape and cannot fundamentally reconstruct the path topology. Furthermore, existing algorithms generally ignore the embedded modeling of engineering vehicle kinematics constraints, resulting in a disconnect between the planning results and the actual control capabilities of the vehicle. SUMMARY

[0005] The application provides a path planning method and system based on connected search inflection points, which solves the problem that the path generated by the prior art cannot adapt to the engineering vehicle working scene.

[0006] To solve the above technical problems, the application provides a path planning method based on connected search inflection points, comprising the following steps:

[0007] Step S1: Obtain the vertex coordinates of each obstacle in the path map, construct an obstacle map according to the vertex coordinates of the obstacle, construct a target set, and store the starting point and endpoint of the path planning as a target point pair in the target set;

[0008] Step S2: Connect all target point pairs in the target set, and for each obstacle intersecting the target point connecting line: calculate the directed distance from all vertices of the obstacle to the target point connecting line, and select the vertex with the maximum absolute value of the directed distance on the positive and negative distance sides of the target point connecting line as the effective vertex of the obstacle; if a target point connecting line does not intersect any obstacle, the corresponding target point pair is deleted from the target set and added to the result set;

[0009] Step S3: Take the sum of the distances of the two effective vertices to the target point connecting line as the weight of the obstacle, take the obstacle with the highest weight as the maximum weight obstacle, calculate the comprehensive cost of the two effective inflection points of the maximum weight obstacle respectively, select the effective inflection point with smaller comprehensive cost as the optimal effective inflection point, take the optimal effective inflection point as the intermediate point, and update the target point pairs in the target set;

[0010] Step S4: Repeat steps S2 to S3 until the target set is empty, connect the target point pairs in the result set, and obtain the path planning result.

[0011] Preferably, the expression for calculating the comprehensive cost of the two effective inflection points of the maximum weight obstacle in step S3 is:

[0012] ;

[0013] ;

[0014] In the above formula, , are the comprehensive costs of the effective inflection points A and B respectively; , are the distances from the A point and the B point to the target point connecting line respectively; , are the numbers of obstacles intersecting the target point connecting line after taking the A point and the B point as the optimal effective inflection points respectively.​ is the length of the line connecting the target points; is the number of obstacles intersecting the line connecting the target points; is the distance deviation influence factor.

[0015] Preferably, after obtaining the path planning result in step S4, the path planning result is locally optimized, including the following steps:

[0016] Step S41: Obtain the ordered result point set corresponding to the result set;

[0017] Step S42: Traverse the ordered result point set from the starting point, if the line connecting the starting point and the non-adjacent point does not intersect with any obstacle, delete all points between the starting point and the non-adjacent point, and continue to traverse the ordered result point set from the next point of the starting point;

[0018] Step S43: Repeat step S42 until the entire ordered result point set is traversed, sequentially connect all points in the ordered result point set to obtain the locally optimized path planning result.

[0019] Preferably, the expression for calculating the directed distance of all vertices of the obstacle to the line connecting the target points in step S2 is:

[0020] ;

[0021] In the formula, is the directed distance to the line connecting the target points; , , are the coordinates of the two target points; is the coordinate of the ith vertex of the nth obstacle intersecting the line connecting the target points.

[0022] The application also provides a path planning system based on line search inflection point, which is realized based on the path planning method based on line search inflection point, and includes an initialization module, a data set updating module and a path generation module.

[0023] The initialization module: obtains the coordinates of the starting point and the ending point and the vertex coordinates of all obstacles, constructs an obstacle map, initializes the target set as {(starting point, ending point)}, and initializes the result set.

[0024] ​​​The dataset update module iterates through each pair of target points in the target set. For each pair of target points, it checks whether the line connecting the target points intersects with any obstacle. If there is no intersection, the pair of points is removed from the target set and added to the result set. If there is an intersection, it calculates the directed distances from all vertices of each intersecting obstacle to the line connecting the target point. The vertex with the largest absolute value of the directed distance on both the positive and negative distance sides is selected as the valid vertex. The effective inflection point with the smaller overall cost is selected as the optimal effective inflection point, and the optimal effective inflection point is added to the target set.

[0025] The path generation module updates the target set and result set using the dataset update module until the target set is empty, then connects the point pairs in the result set to form a continuous path.

[0026] Preferably, the system further includes a local optimization module, which: starting from the starting point of the continuous path, sequentially checks the connections between non-adjacent points; if the connection does not intersect with any obstacle, then deletes all points between the starting point and non-adjacent points, and continues to check the intersection of the connections between non-adjacent points and obstacles starting from the next point that has not been deleted, until the entire path has been traversed.

[0027] Preferably, in the dataset update module, the expression for calculating the overall cost is:

[0028] ;

[0029] ;

[0030] In the above formula, , These are the combined costs of effective inflection points A and B, respectively. , These are the distances from the lines connecting points A and B to the target point, respectively. , These represent the number of obstacles that intersect the line connecting points A and B to the new target point after taking points A and B as the optimal effective inflection points; The length of the line connecting the target points; The number of obstacles that intersect the line connecting to the target point; This is the distance deviation influencing factor.

[0031] Preferably, the system further includes a debugging module, which adjusts the deviation influence factor based on the number of obstacles in the map, the map area, and planning preferences. The size should be increased when obstacles are dense, numerous, and occupy a large area of ​​the map. When obstacles are sparse, few in number, and occupy a small area of ​​the map, adjust the size. When the planning preference is to optimize the number of path turns, increase the value. When the planning preference is to optimize the total path distance, reduce the value. .

[0032] The application further provides an electronic device, comprising a memory, a processor and a computer program, the computer program being stored in the memory and configured to be executed by the processor to implement the path planning method based on the connected search inflection point.

[0033] The application further provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the path planning method based on the connected search inflection point.

[0034] The application has at least the following beneficial effects:

[0035] 1. The obstacle map is constructed by acquiring the vertex coordinates of the obstacles, and the vertex with the maximum absolute value of the directed distance is selected as the effective vertex of the obstacle on the positive and negative distance sides of the target point connection, so that the planned path can closely fit the boundary of the obstacle, and the problem that the path is too conservative in the traditional algorithm is avoided.

[0036] 2. The path is more smooth by selecting the optimal effective inflection point as the intermediate point and constantly updating the target point pairs in the target set, the number of turning points and the total turning angle in the path are greatly reduced, the load of the steering mechanism of the engineering vehicle is reduced, the tire wear is reduced, and the stability and safety of the vehicle driving are improved. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 The figure is a method flowchart of the embodiment of the application;

[0038] Figure 2 The figure is a grid map of the embodiment of the application;

[0039] Figure 3 The figure is an intersection diagram of the embodiment of the application with the obstacle;

[0040] Figure 4 The figure is an intersection diagram of the second-level connection , with the obstacle in the embodiment of the application;

[0041] Figure 5 The figure is a diagram for finding the next-level connection in the embodiment of the application;

[0042] Figure 6 The figure is a diagram of the final path generated in the embodiment of the application;

[0043] Figure 7 The figure is a performance diagram of the path planning algorithm in the embodiment of the application;

[0044] Figure 8 For the traditional performance diagram of the algorithm;

[0045] Figure 9 For the improved performance diagram of the algorithm;

[0046] Figure 10 For the implementation effect diagram of the path planning algorithm of the embodiment of the application under various working conditions;

[0047] Figure 11 For the planning result comparison diagram of the method of the embodiment of the application and various path planning methods;

[0048] Figure 12 For the system structure diagram of the embodiment of the application. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the application will be apparently and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments of the application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the application.

[0050] As Figure 1 shown, the embodiment of the application provides a path planning method based on a turning point of a connection line (SETL, Start-End Turningpoints Linebased Path Algorithm), including the following steps:

[0051] Step S1: Obtain the vertex coordinates of each obstacle in a path map, construct an obstacle map according to the vertex coordinates of the obstacle, construct a target set, and store the start point and the end point of path planning as a target point pair in the target set.

[0052] Specifically, first, the target points of path planning are determined, and there can be multiple target points. In the embodiment of the application, the start point and the end point are taken as two target points of the first connection line. The start point is taken as the start point, the end point is taken as the end point, and the constructed target point set is . . .

[0053] Step S2: Connect all target point pairs in the target set, and for each obstacle intersected by the target point connection line: calculate the directed distance from all vertices of the obstacle to the target point connection line, and select the vertex with the maximum absolute value of the directed distance on the positive and negative distance sides of the target point connection line as the effective vertex of the obstacle; if the target point connection line is not intersected by the obstacle, the obstacle is not considered in the path planning process. not intersect with any obstacle, and The corresponding target point pair is deleted from the target set and added to the result set.

[0054] Specifically, the connection with is obtained. Obstacles intersecting with the line segment are identified, and the two vertices on each obstacle that are farthest from the line segment are found as the effective turning points of the obstacle.

[0055] , are the coordinates of the start point and the end point , respectively, so the equation of the line segment can be obtained:

[0056] .

[0057] Let the coordinates of the i-th vertex of the n-th obstacle intersecting with be , and the distance from the vertex to be . The expression for is:

[0058] .

[0059] The two vertices of each obstacle that are farthest from are taken as the effective turning points.

[0060] Step S3: The sum of the distances from the two effective vertices to the target point is taken as the weight of the obstacle, the obstacle with the highest weight is taken as the maximum weight obstacle, the comprehensive cost of the two effective turning points of the maximum weight obstacle is calculated respectively, the effective turning point with the smaller comprehensive cost is selected as the optimal effective turning point, the optimal effective turning point is taken as the intermediate point, and the target point pair in the target set is updated.

[0061] Specifically, the sum of the distances from the two effective vertices to is taken as the weight of the obstacle, the obstacle with the highest weight is taken as the maximum weight obstacle of , the comprehensive cost of the two effective turning points of the maximum weight obstacle is calculated, the effective turning point with the smaller comprehensive cost is selected as the first optimal effective turning point , and the target point set is updated to .

[0062] In this embodiment of the invention, when selecting the optimal effective inflection point, the two effective inflection points of the obstacle with the highest weight at that level are selected. If only the effective inflection point with the smaller distance is used as the optimal effective inflection point, the selected optimal effective inflection point may not be the global optimal, resulting in the features of the next level connection between the two effective inflection points and the obstacle not being used well.

[0063] Based on this characteristic, the embodiments of the present invention additionally introduce a distance deviation influence factor. ,pass Comprehensive cost The calculation is made using a non-linear approach. Distance deviation influence factor. The distance deviation referred to in the text is the distance deviation between the two effective inflection points of the highest-weight obstacle in that level of connection, i.e. When the distance deviation is less than a certain percentage of the length of the cascade, right The influence weight of the value will decrease, making it even more necessary to... The number of obstacles that intersect when used as the target point for the next level of connection. The influence of distance deviation on the length of the connecting line is used to determine which point, A or B, is better. When the distance deviation exceeds a certain percentage of the line length, right The influence weight of the value will increase, at which point the number of intersections between the level connection and obstacles will be added. Thus dilute The number of obstacles that intersect when used as the target point for the next level of connection. The influence of distance deviation. This proportion is the distance deviation influence factor. By introducing this distance deviation influence factor This allows for a more accurate selection of the optimal effective inflection point, resulting in the optimal overall path planning, i.e., the optimal combination of path length and turning conditions.

[0064] Specifically, calculation and The two valid inflection points of each intersecting obstacle to distance and ,Will The corresponding obstacles are The maximum weighted obstacle. Calculate the combined cost of the two valid inflection points of the maximum weighted obstacle, denoted as points A and B respectively, and calculate their combined costs. :

[0065] ;

[0066] ;

[0067] In the above formula, , the total cost of A and B points respectively; , the distance of A and B points respectively to ; , the number of obstacles intersected by the line connecting A and B points respectively as the next level target points; , the length of the current line segment ; the number of obstacles intersected by ; the distance deviation influence factor, the greater, the less the total steering angle of the steering point, the longer the relative distance, the smaller, the shorter the distance, the more the total steering angle of the steering point.

[0068] Compare the total cost and select the optimal effective inflection point of this level as :

[0069] .

[0070] The distance deviation influence factor can be adjusted according to the working conditions. When the obstacles are relatively dense, the number is large and occupies a large area of the map, the distance deviation influence factor can be appropriately increased to about ; when the obstacles are relatively sparse, the number is small and occupies a small area of the map, the distance deviation influence factor can be appropriately reduced to about . It can also be adjusted according to the planning preference of steering. When the planning preference is steering, the distance deviation influence factor can be appropriately increased; when the planning preference is shorter distance, the distance deviation influence factor can be appropriately reduced. According to a large amount of experimental data analysis, when the distance deviation influence factor , the comprehensive performance index is optimal.

[0071] Step S4: Repeat steps S2 to S3 until the target set is empty, connect the target points in the result set to obtain the path planning result.

[0072] Specifically, after adding the optimal effective inflection point of each level line to the target point set, the next level line is performed, and then the search is performed layer by layer, i.e. steps S3-S5 are repeatedly executed by level. If the line of a certain pair of target points does not intersect with any obstacle, the pair of target points is deleted from the target point set, and the target point is stored in the result set until the target point set in step S1 is empty, i.e. there is no pair of target points that need to search for the corresponding optimal effective inflection point, i.e. all results are found. According to the vertices in the result set, a preliminary path before local optimization is generated.​

[0073] Step S7: Starting from the starting point, check all vertices in the preliminary path that are not adjacent to the starting point. If the line connecting the starting point and a certain vertex does not intersect any obstacle, delete all intermediate points between the starting point and that vertex, and keep the line as a new path to optimize the preliminary path.

[0074] Specifically, the initial path is locally optimized. Starting from the starting point, each vertex is connected sequentially. If the line connecting a vertex to a non-adjacent vertex does not intersect any obstacle, then all vertices between the vertex and that non-adjacent vertex are deleted, and the connection is retained as the new path.

[0075] For example, if the initial path without local optimization is:

[0076] .

[0077] Get the result set .from Begin by connecting to all subsequent non-adjacent vertices in sequence, i.e., connecting to all vertices except... Connect the vertices in sequence. If a certain connection, for example... Clear if it does not intersect with any obstacles. and All optimal effective inflection points between, i.e. Then, starting from the next point that has not been deleted, continue connecting it to all subsequent vertices except for adjacent ones. Delete vertices that satisfy the above conditions to obtain the locally optimized result set. Connect each line segment in the result set to obtain the final locally optimized path:

[0078] .

[0079] Complete the path planning based on the path planning method of inflection points in a linear search.

[0080] Example 1

[0081] For example Figure 2 As shown, the size is Taking a grid map G as an example, firstly, the vertex coordinates of the grid map G are extracted to obtain the obstacle coordinate system map:

[0082] .

[0083] The current target point set is ,in , obtain the first level connection The equation:

[0084] ;

[0085] ;

[0086] .

[0087] As Figure 3 shown, the line and the obstacle coordinate system map have 9 intersections with the obstacles, the distance of each obstacle vertex to the line is calculated :

[0088] .

[0089] Find the two vertices on both sides of each obstacle with the maximum distance as the effective inflection point, calculate the distance sum of the two effective inflection points, find the corresponding obstacle as the maximum weight obstacle of the line. The maximum weight obstacle in this embodiment is the 7th obstacle in , and the coordinates of its two effective inflection points are Figure 3 .

[0090] Set , calculate the comprehensive cost of the two effective inflection points. At present , , , , , substitute into the comprehensive cost calculation formula to obtain . Therefore, select as the first level optimal effective inflection point .

[0091] Store the first level optimal effective inflection point in the target point set to obtain . Repeat the above steps of finding the optimal effective inflection point to obtain the second level line , . As Figure 4 shown, find the second level optimal effective inflection point , and continue to search for the line downward, as Figure 5 shown.

[0092] If the line of a paired target point does not intersect with any obstacle, delete the target point in the target point set and store it in the result set. Until the target point set is empty, that is, there is no paired target point that needs to search for the corresponding optimal effective inflection point, that is, all the results are found, and the result set is obtained, which is the preliminary path before local optimization.

[0093] ​The initial path is locally optimized, starting from the starting point, connecting each vertex in turn, if the connection between the starting point and a non-adjacent vertex does not intersect with any obstacle, all the vertices between the connection are deleted, and the connection is reserved as a new path. In this embodiment, there is no path that needs to be locally optimized, so the initial path before local optimization is directly used as the final path. As shown in Figure 6 The final path after the entire path planning is completed is shown in

[0094] The path planning algorithm is one of the most commonly used heuristic search algorithms in path planning, and its core idea is to select the optimal path by comprehensively considering the actual cost g(n) of the node and the heuristic estimation cost h(n). The core formula of the path planning algorithm is f(n)=g(n)+h(n), wherein g(n) represents the actual cost from the starting point to the current node, and h(n) represents the estimated cost from the current node to the target node. The path planning algorithm manages nodes by maintaining an OPEN table and a CLOSE table, and preferentially expands the node with the smallest f(n) value until the target node is found or the OPEN table is empty. Subsequent researchers have proposed various optimization methods for the path planning algorithm. For example, the bidirectional algorithm searches from the starting point and the end point at the same time, significantly reducing the search range and improving the efficiency. The dynamic weighting method uses a larger heuristic function weight at the beginning of the search to quickly approach the target, and reduces the weight at the later stage to accurately search for the optimal path. In addition, the jump point search technology detects the blocked points on the straight path, and directly jumps over these points, thereby reducing the search space and improving the search efficiency.

[0095] The path planning method of the embodiment of the present application is compared with the traditional algorithm and the improved algorithm, and the path planning result is obtained as shown in Figures 7-9 As shown in Figure 7 , the total path length obtained by the embodiment is 60.2454, the number of turns is 13, the total turning angle is 451.252, and the planning time is 0.0181s. As shown in Figure 8 , the total path length obtained by using the traditional algorithm is 63.9411, the number of turns is 26, the total turning angle is 1170, and the planning time is 0.0786s. As shown in Figure 9 , the total path length obtained by using the improved algorithm is 68.7668, the number of turns is 14, the total turning angle is 657.0598, and the planning time is 0.100168s. The method of the embodiment of the present application is compared with the traditional algorithm and the improved Compared to other algorithms, this method optimizes path routing using obstacle vertex features, reducing the total distance by 8%-15%, especially in situations with dead-end obstacles, avoiding redundant detours. The smoother steering angles improve path continuity, reducing the total steering angle by 50%-70%, resulting in more stable vehicle operation and optimized energy consumption. In vector maps such as CAD architectural floor plans, the planning success rate of the method in this embodiment is higher than... The algorithm is 32% faster, and the computation time is not affected by map type. The rasterization process of the algorithm will result in an additional 15%-20% time overhead. By dynamically selecting the obstacle with the highest weight and the optimal inflection point, the method of this embodiment significantly improves the avoidance efficiency of U-shaped and maze-shaped obstacles. In simulation tests, the method of this embodiment achieved a first-time planning success rate of up to 98% for dead-end obstacles, while... The algorithm's success rate is only 78% due to backtracking from invalid nodes.

[0096] like Figure 10 The diagram shows the path planning results of the method of this invention under various working conditions. It can be seen that the method of this invention exhibits superior planning performance under various working conditions, indicating that the method of this invention is adaptable to path planning under various complex working conditions. Specifically, the method of this invention utilizes innovative mechanisms such as steering cost function reconstruction, dynamic weighting of obstacle vertices, and continuous space search to simultaneously optimize the number of steering points, total steering angle, path length, and computational efficiency during the path generation stage. By reducing the number of steering points and the total steering angle, the stability of the engineering vehicle can be significantly improved: Firstly, the number of steering points is reduced by more than 50%, reducing the need for frequent vehicle attitude adjustments, thereby avoiding lateral swaying caused by centroid shift; secondly, the total steering angle is compressed to the 400-600° range, reducing the inertial impact and hydraulic system pressure fluctuations during a single steering maneuver, ensuring a smooth and controllable steering process; thirdly, combined with continuous curvature path design, it further suppresses vehicle body vibration and tire slippage. Data shows that loaders using this method experience a 35%-45% reduction in lateral acceleration fluctuations and a 60% decrease in the incidence of abnormal vibrations in the steering mechanism in typical mining scenarios. Simultaneously, the increased handling tolerance due to improved stability reduces the frequency of emergency braking by over 40%. This breakthrough not only alleviates efficiency losses and mechanical wear caused by frequent steering in construction vehicles but also provides key technological support for new operational models such as reduced manpower in mines and zero emissions on construction sites by shortening paths and responding to dynamic environments in real time. This accelerates the evolution of the construction machinery industry towards intelligence and sustainability.

[0097] like Figure 11The figure shows a comparison of the planning results of the method of this invention and various path planning methods. GA represents the Genetic Algorithm, SSA the Sparrow Search Algorithm, PSO the Particle Swarm Optimization Algorithm, GWO the Grey Wolf Algorithm, AO the Eagle Algorithm, IAO the Information Acquisition Algorithm, and DBO the Dung Beetle Algorithm. As can be seen from the figure, the method of this invention, by replacing node-by-node expansion with a linear search, significantly reduces invalid computation. In a grid map of the same size, the average planning time and the number of path turning points of the method of this invention are significantly reduced, and the efficiency advantage is more pronounced the larger the map.

[0098] like Figure 12 The diagram shows a structural schematic of a path planning system based on a linear search inflection point according to an embodiment of the present invention, including: an initialization module, a dataset update module, a path generation module, a local optimization module, and a debugging module.

[0099] Initialization module: Obtain the coordinates of the start point, the end point, and the vertex coordinates of all obstacles, construct an obstacle map, initialize the target set as {(start point, end point)}, and initialize the result set.

[0100] Dataset update module: Iterate through each pair of target points in the target set. For each pair of target points, check whether the line connecting the target points intersects with any obstacle. If there is no intersection, remove the pair of points from the target set and add it to the result set. If there is an intersection, calculate the directed distance from all vertices of each intersecting obstacle to the line connecting the target point. Select the vertex with the largest absolute value of the directed distance on both the positive and negative distance sides as the valid vertex. Select the valid inflection point with the smaller overall cost as the optimal valid inflection point and add the optimal valid inflection point to the target set.

[0101] Path generation module: Updates the target set and result set using the dataset update module until the target set is empty, then connects the point pairs in the result set to form a continuous path.

[0102] Local optimization module: Starting from the beginning of the continuous path, check the connection between the point and the non-adjacent point in turn. If the connection does not intersect with any obstacle, delete all points between the starting point and the non-adjacent point. Start from the next point that has not been deleted and continue to check the intersection of the connection between the non-adjacent point and the obstacle, until the entire path is traversed.

[0103] Debugging module: Adjusts deviation influencing factors based on the number of obstacles in the map, map area, and planning preferences. The size should be increased when obstacles are dense, numerous, and occupy a large area of ​​the map. When obstacles are sparse, few in number, and occupy a small area of ​​the map, adjust the size. When the planning preference is to optimize the number of path turns, increase the value. When the planning preference is to optimize the total path distance, reduce the value. .

[0104] The embodiment of the present application further provides an electronic device, comprising a memory, a processor and a computer program, the computer program is stored in the memory and is configured to be executed by the processor to realize the path planning method based on the connected search inflection point.

[0105] The embodiment of the present application further provides a computer readable storage medium, the computer readable storage medium stores a computer program, and the computer program is executed by a processor to realize the path planning method based on the connected search inflection point.

[0106] The technical features of the above embodiments can be combined in any manner. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, only the preferred embodiments of the present application are expressed, and the description is more specific and detailed, but it cannot be understood as a limitation on the scope of the present application. As long as the combination of these technical features does not exist, it should be considered as the scope of the present application.

[0107] It should be pointed out that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for path planning based on connected search inflection points, characterized in that, The method comprises the following steps: Step S1: Obtain the vertex coordinates of each obstacle in the path map, construct an obstacle map according to the vertex coordinates of the obstacles, construct a target set, and store the start point and end point of the path planning as a target point pair in the target set; Step S2: Connect all target point pairs in the target set, and for each obstacle intersecting the target point connecting line: calculate the directed distance from all vertices of the obstacle to the target point connecting line, and select the vertex with the maximum absolute value of the directed distance on the positive and negative distance sides of the target point connecting line as the effective vertex of the obstacle; If a goal point line does not intersect with any obstacle, The corresponding goal point pair is deleted from the goal set and added to the result set; Step S3: Take the sum of the distances of the two effective vertices to the target point connecting line as the weight of the obstacle, take the obstacle with the highest weight as the maximum weight obstacle, calculate the comprehensive cost of the two effective inflection points of the maximum weight obstacle respectively, select the effective inflection point with the smaller comprehensive cost as the optimal effective inflection point, take the optimal effective inflection point as an intermediate point, and update the target point pair in the target set; Step S4: Repeat steps S2 to S3 until the target set is empty, connect the target point pairs in the result set, and obtain the path planning result.

2. The path planning method based on the connected search inflection point according to claim 1, characterized in that: The expression for calculating the comprehensive cost of the two effective inflection points of the maximum weight obstacle in step S3 is: ； ； In the above formula, , are the comprehensive cost of effective inflection points A and B, respectively; , are the distances from the A point and the B point to the target point line, respectively; , are the number of obstacles intersecting the new target point line after taking the A point and the B point as the optimal effective inflection points, respectively; is the length of the current target point line; is the number of obstacles intersecting the target point line; is the distance deviation influence factor.

3. The method of claim 1, wherein: After obtaining the path planning result in step S4, the path planning result is locally optimized, comprising the following steps: Step S41: Obtain the ordered result point set corresponding to the result set; Step S42: From the start point, traverse the ordered result point set. If the line connecting the start point and the non-adjacent point does not intersect with any obstacle, delete all points between the start point and the non-adjacent point, and continue to traverse the ordered result point set from the next point of the start point. Step S42: From the start point, traverse the ordered result point set. If the line connecting the start point and the non-adjacent point does not intersect with any obstacle, delete all points between the start point and the non-adjacent point, and continue to traverse the ordered result point set from the next point of the start point. <000 Step S43: Repeat step S42 until the entire ordered result point set is traversed, sequentially connect all points in the ordered result point set, and obtain the locally optimized path planning result.

4. The method of claim 1, wherein: The expression for calculating the directed distance from all vertices of the obstacle to the target point connecting line in step S2 is: ； In the formula, is a directed distance to the connecting line of the target points; , is the coordinates of the two target points; is the coordinates of the i-th vertex of the n-th obstacle intersecting the connecting line of the target points.

5. A path planning system based on the connected search turning point, which is implemented based on the path planning method based on the connected search turning point according to any one of claims 1 to 4, characterized in that, It comprises: An initialization module, a data set updating module, and a path generation module; The initialization module: obtains the coordinates of the start point and the end point and the vertex coordinates of all obstacles, constructs an obstacle map, initializes the target set as {(start point, end point)}, and initializes the result set; The data set updating module: traverses each target point pair in the target set, detects whether the connecting line of the target point pair intersects any obstacle for each target point pair, removes the point pair from the target set and adds it to the result set if there is no intersection; If there is an intersection, calculate the directed distance from all vertices of each intersecting obstacle to the target point connecting line, select the vertex with the maximum absolute value of the directed distance on the positive and negative distance sides as the effective vertex, and select the optimal effective inflection point with the smaller comprehensive cost from the effective inflection points and add it to the target set; The path generation module: updates the target set and the result set using the data set updating module until the target set is empty, connects the point pairs in the result set, and forms a continuous path.

6. The system according to claim 5, wherein: The system further comprises a local optimization module, which: starts from the start point of the continuous path and sequentially checks the connecting line of the non-adjacent point, deletes all points between the start point and the non-adjacent point if the connecting line does not intersect any obstacle, and continues to check the intersection of the connecting line of the non-adjacent point and the obstacle from the next point that has not been deleted until the entire path is traversed.

7. The system of claim 5, wherein: In the data set updating module, the expression for calculating the comprehensive cost is: ； ； In the above formula, , are the comprehensive cost of effective inflection points A and B, respectively; , are the distances from the A point and the B point to the target point line, respectively; , are the number of obstacles intersecting the new target point line after taking the A point and the B point as the optimal effective inflection points, respectively; is the length of the current target point line; is the number of obstacles intersecting the target point line; is the distance deviation influence factor.

8. The system according to claim 7, wherein: The system further comprises a debugging module, which adjusts the size of the deviation influence factor according to the number of obstacles in the map, the area of the map, and planning preferences ; when the obstacles are dense, numerous, and occupy a large area of the map, the deviation influence factor is increased ; when the obstacles are sparse, few, and occupy a small area of the map, the deviation influence factor is decreased ; When the planning preference is the number of path turning times optimization, increase ; When the planning preference is path total distance optimization, turn down .

9. An electronic device comprising: The memory, the processor and the computer program are characterized in that the computer program is stored in the memory and is configured to be executed by the processor to realize the path planning method based on the connected search inflection point according to any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that: The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the path planning method based on the connected search inflection point according to any one of claims 1 to 4.

Images