Robot motion planning method based on greedy algorithm and improved mixed A* algorithm
By using a greedy algorithm and an improved hybrid A* algorithm, the wall-climbing robot generates the shortest safe path that satisfies kinematic constraints in a large, enclosed space, solving the problems of path planning and trajectory tracking, and improving detection efficiency and safety.
Patent Information
- Application Number
- CN202410277403.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-12
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-03-12
AI Technical Summary
Existing wall-climbing robots face challenges in path planning and trajectory tracking within large, enclosed spaces, making them unable to effectively avoid obstacles, resulting in low detection efficiency and potential safety hazards.
A greedy algorithm is used to determine the traversal order of the detection area, and an improved hybrid A* algorithm is combined for path planning. Taking into account the robot's attitude angle and environmental constraints, the shortest safe path that satisfies the kinematic constraints is generated.
This improved the accuracy and safety of path planning for wall-climbing robots, reduced safety accidents caused by obstacles and attitude angles, and ensured the safe operation and production efficiency of the system.
Smart Images

Figure CN118960729A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of robot technology, and in particular to a robot motion planning method based on a greedy algorithm and an improved hybrid A* algorithm. Background Art
[0002] Most large confined spaces have complex structures and are subject to difficult industrial scenarios such as weak light, high dust content, and magnetic sealing. Taking the industrial inspection of the heating surface of a large boiler as an example, it currently mainly relies on manual operations such as hand touch and visual inspection on the scaffolding. The missed detection rate is high and the efficiency is low. It is difficult to achieve a high coverage rate for the furnace tube inspection during each boiler maintenance, which is an important reason for the frequent boiler accidents. In addition, high-altitude operations on scaffolding that rely on manual inspection under harsh conditions often cause a large number of personnel safety accidents and accidental damage to furnace tubes.
[0003] Accessing a large confined space is a complex and potentially dangerous task that requires careful planning and execution. The specific steps for accessing a large confined space may vary depending on the nature of the task and the characteristics of the equipment. Before performing any work, ensure that local regulations and safety standards are met and that personnel are properly trained. Safety is always the primary consideration when accessing a large confined space. Ensure that all steps are carefully followed to reduce potential risks and hazards.
[0004] Replacing inefficient manual inspection operations in harsh environments with wall-climbing robots can effectively reduce personnel safety accidents and accidental damage to furnace tubes, improve system safety and reliability, ensure safe system operation, ensure safe production, and improve industrial economic benefits. However, due to the non-complete characteristics of the wall-climbing robot itself and the special operating scenarios of the water-cooled wall tubes, the wall-climbing robot cannot move in the vertical direction of the current head direction, that is, it cannot move horizontally, and the change of its attitude angle needs to be achieved through robot displacement. When coking, manhole doors and other inaccessible areas appear on the water-cooled wall surface, the existing wall-climbing robot cannot perform good obstacle avoidance control.
[0005] Therefore, how to propose a technical solution for path planning and trajectory tracking of wall-climbing robots has become a problem that needs to be solved urgently. Summary of the invention
[0006] Technical issues to be solved
[0007] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a robot motion planning method based on a greedy algorithm and an improved hybrid A* algorithm, which solves the technical problems of path planning and trajectory tracking of a wall-climbing robot.
[0008] Technical Solution
[0009] In order to achieve the above object, the main technical solutions adopted by the present invention include:
[0010] In a first aspect, the present invention provides a robot motion planning method based on a greedy algorithm and an improved hybrid A* algorithm, comprising:
[0011] A greedy algorithm is used to determine the traversal order of the wall-climbing robot to traverse each detection area, which specifically includes:
[0012] Select a detection area as a starting point;
[0013] Select the detection area that is closest to the current detection area and has not been traversed as the next detection area;
[0014] The next selected detection area is added to the path and marked as visited. The detection area closest to the current detection area and not traversed is selected as the next detection area until all detection areas are marked as visited, and the traversal order is determined after returning to the starting point.
[0015] According to the traversal order, an improved hybrid A* algorithm is used to plan paths between adjacent detection areas.
[0016] Optionally, the adopting of an improved hybrid A* algorithm to perform path planning between adjacent detection areas according to the traversal order includes:
[0017] Initialize the open list and the closed list, where the open list is used to store the nodes to be traversed, and the closed list is used to store the nodes that have been traversed;
[0018] Add the starting point to the open list and set it as the current node, and determine whether the open list is empty; when the open list is not empty, select the node with the lowest f value from the open list as the current node, and plan the path based on whether the current node is the target node; when the open list is empty, no path can be found; where f = g + h, g is the actual cost from the starting point to the current node, and h is the estimated cost from the current node to the target node.
[0019] Optionally, performing path planning according to whether the current node is a target node includes:
[0020] If the current node is the target node, a path is generated according to the path approximation principle, and the path generation is completed;
[0021] If the current node is not the target node, according to the constraint of the body attitude angle θ, traverse all neighboring nodes that meet the constraint, and perform path planning based on whether the path between the current node and the neighboring node passes through obstacles.
[0022] Optionally, the performing path planning according to whether a path between the current node and a neighboring node passes through an obstacle includes:
[0023] If the path between the current node and the neighboring node passes through an obstacle, the neighboring node will be deleted;
[0024] If the path between the current node and the neighboring node does not pass through obstacles, the path is planned based on whether the neighboring node is in the closed list.
[0025] Optionally, performing path planning according to whether the neighbor node is in a closed list includes:
[0026] If the neighbor node is in the closed list, delete the node;
[0027] If the neighbor node is not in the closed list, calculate the g value of the neighbor node to determine whether the neighbor node has ever been in the open list;
[0028] If the neighbor node was once in the open list, recalculate the g value of the neighbor node and determine the g value of the neighbor node based on the current g value and historical g value of the neighbor node;
[0029] If the neighbor node is not in the open list, calculate the h value and f value of the neighbor node, add the neighbor node to the open list, and set the parent node of the neighbor node as the current node.
[0030] Optionally, determining the g value of the neighbor node according to the current g value and the historical g value of the neighbor node includes:
[0031] If the current g value of the neighbor node is less than the historical g value, update the g of the neighbor node to be equal to the current g value;
[0032] If the current g value of the neighbor node is greater than or equal to the historical g value, keep the historical g value.
[0033] Optionally, the method further comprises:
[0034] Control the robot to operate in each inspection area according to the planned path.
[0035] Optionally, the method further comprises:
[0036] During the robot operation, the robot position, posture, image, speed, voltage and current information are fed back in real time;
[0037] Determine whether to change the operation mode or return based on the robot's position, posture, image, speed, and voltage and current information.
[0038] In a second aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm as described in any one of the first aspects above.
[0039] In a third aspect, the present invention provides a storage device comprising a storage medium and a processor, wherein the storage medium stores a computer program, and when the program is executed by the processor, the robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm as described in any one of the first aspects above is implemented.
[0040] Beneficial Effects
[0041] The beneficial effects of the present invention are: a method based on a greedy algorithm and an improved hybrid A*
[0042] The robot motion planning method of the algorithm uses a greedy algorithm and an improved hybrid A* algorithm to plan the path of the wall-climbing robot, which can provide the wall-climbing robot with a shortest safe path that meets the kinematic constraints and can be tracked by the wall-climbing robot. At the same time, the rasterization of the environmental map can improve the precision and accuracy of the path planning of the wall-climbing robot. The wall-climbing robot can work according to the planning results to reduce the probability of injuries caused by coking, manhole door slips, crashes, etc. during the movement of the wall-climbing robot, which can effectively reduce personnel safety accidents, improve system safety and reliability, ensure the safe operation of the system, ensure safe production, and improve the economic benefits of the industry. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Figure 1 A schematic diagram of a flow chart of a robot motion planning method based on a greedy algorithm and an improved hybrid A* algorithm provided in an embodiment of the present invention;
[0044] Figure 2 A diagram of a case implementation process based on a greedy algorithm and an improved hybrid A* algorithm provided by another embodiment of the present invention;
[0045] Figure 3 A schematic diagram of a greedy algorithm provided in another embodiment of the present invention for solving the sorting problem of multiple areas to be detected;
[0046] Figure 4 A schematic diagram of a path generated by an improved hybrid A* provided in another embodiment of the present invention. DETAILED DESCRIPTION
[0047] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.
[0048] This embodiment aims at the working scenario of a wall-climbing robot inspecting the water-cooled wall of a boiler in a large enclosed boiler space, and solves the path planning and trajectory tracking problems of the wall-climbing robot. To solve this problem, we must first explore the path planning algorithm of the wall-climbing robot and design a planning algorithm that can meet the model and solve the limitations in the current field. According to the known environmental map, a shortest safe path that meets the kinematic constraints and can be tracked by the wall-climbing robot is planned for the wall-climbing robot, which reduces the possibility of the wall-climbing robot crashing, enhances the safety of the operation, and reduces the energy consumption of the wall-climbing robot operation, further reducing the cost of the wall-climbing robot to inspect the water-cooled wall. By using a greedy algorithm and an improved hybrid A* algorithm, a safe path that can be designed for the wall-climbing robot that traverses all the areas to be detected without repetition and can be stably tracked can be designed.
[0049] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.
[0050] First, refer to Figure 1 This embodiment provides a robot motion planning method based on a greedy algorithm and an improved hybrid A* algorithm, including:
[0051] S1, a greedy algorithm is used to determine the traversal order of the wall-climbing robot to traverse each detection area.
[0052] S1 specifically includes:
[0053] S11, select a detection area as a starting point.
[0054] S12, selecting a detection area that is closest to the current detection area and has not been traversed as the next detection area.
[0055] S13, adding the next selected detection area to the path and marking the detection area as visited, continuing to select the detection area closest to the current detection area and not traversed as the next detection area, until all detection areas are marked as visited, returning to the starting point, and the traversal order is determined.
[0056] S2, using the improved hybrid A* algorithm to plan paths between adjacent detection areas according to the traversal order.
[0057] In this embodiment, the starting point of the planning is the current detection area, and the target point is the next detection area. When performing path planning, the map will be rasterized between the starting point and the target point to form a map node.
[0058] Optionally, the adopting of an improved hybrid A* algorithm to perform path planning between adjacent detection areas according to the traversal order includes:
[0059] Initialize the open list and the closed list, where the open list is used to store the nodes to be traversed, and the closed list is used to store the nodes that have been traversed;
[0060] Add the starting point to the open list and set it as the current node, and determine whether the open list is empty; when the open list is not empty, select the node with the lowest f value from the open list as the current node, and plan the path based on whether the current node is the target node; when the open list is empty, no path can be found; where f = g + h, g is the actual cost from the starting point to the current node, and h is the estimated cost from the current node to the target node.
[0061] Optionally, performing path planning according to whether the current node is a target node includes:
[0062] If the current node is the target node, a path is generated according to the path approximation principle, and the path generation is completed;
[0063] If the current node is not the target node, according to the constraint of the body attitude angle θ, traverse all neighboring nodes that meet the constraint, and perform path planning based on whether the path between the current node and the neighboring node passes through obstacles.
[0064] Optionally, the performing path planning according to whether a path between the current node and a neighboring node passes through an obstacle includes:
[0065] If the path between the current node and the neighboring node passes through an obstacle, the neighboring node will be deleted;
[0066] If the path between the current node and the neighboring node does not pass through obstacles, the path is planned based on whether the neighboring node is in the closed list.
[0067] Optionally, performing path planning according to whether the neighbor node is in a closed list includes:
[0068] If the neighbor node is in the closed list, delete the node;
[0069] If the neighbor node is not in the closed list, calculate the g value of the neighbor node to determine whether the neighbor node has ever been in the open list;
[0070] If the neighbor node was once in the open list, recalculate the g value of the neighbor node and determine the g value of the neighbor node based on the current g value and historical g value of the neighbor node;
[0071] If the neighbor node is not in the open list, calculate the h value and f value of the neighbor node, add the neighbor node to the open list, and set the parent node of the neighbor node as the current node.
[0072] Optionally, determining the g value of the neighbor node according to the current g value and the historical g value of the neighbor node includes:
[0073] If the current g value of the neighbor node is less than the historical g value, update the g of the neighbor node to be equal to the current g value;
[0074] If the current g value of the neighbor node is greater than or equal to the historical g value, keep the historical g value.
[0075] Optionally, the method further comprises:
[0076] Control the robot to operate in each inspection area according to the planned path.
[0077] Optionally, the method further comprises:
[0078] During the robot operation, the robot position, posture, image, speed, voltage and current information are fed back in real time;
[0079] Determine whether to change the operation mode or return based on the robot's position, posture, image, speed, and voltage and current information.
[0080] The present invention discloses a robot motion planning method based on a greedy algorithm and an improved hybrid A* algorithm. The greedy algorithm and the improved hybrid A* algorithm are used to plan the path of a wall-climbing robot, and a shortest safe path that satisfies kinematic constraints and can be tracked by the wall-climbing robot can be provided. At the same time, the rasterization of the environmental map can improve the precision and accuracy of the path planning of the wall-climbing robot. The wall-climbing robot operates according to the planning results, which can reduce the probability of injuries and other incidents caused by coking, manhole door slips, crashes, etc. during the movement of the wall-climbing robot, effectively reduce personnel safety accidents, improve system safety and reliability, ensure safe operation of the system, ensure safe production, and improve the economic benefits of the industry.
[0081] The following is a specific example to further illustrate the robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm of the present invention:
[0082] First, for multiple areas to be inspected, sort them and find a sequence that can traverse all the areas to be inspected with the lowest cost. This problem can be abstracted and considered as a traveling salesman problem. The wall-climbing robot is regarded as a "traveling businessman" and each key inspection area is regarded as N "cities" that the traveling businessman needs to visit. Then the goal is to find an optimal traversal sequence that allows the wall-climbing robot to traverse every key inspection area without omission and return to the starting point of the wall-climbing robot.
[0083] There are many ways to solve the Traveling Salesman Problem (TSP), some of the main ones include:
[0084] Brute Force: This is an intuitive but computationally expensive method. It tries to enumerate all possible paths, calculate their total length, and finally select the shortest path. However, as the number of cities increases, the computational complexity of the brute force method grows exponentially, so it is not practical for large-scale problems.
[0085] Dynamic programming: TSP can be solved by dynamic programming. The dynamic programming algorithm uses a bottom-up approach to gradually calculate the optimal solution to the sub-problems and finally obtain the solution to the overall problem. Although dynamic programming can solve TSP in theory, it still has the problem of high computational complexity for large-scale problems.
[0086] Greedy algorithm: Greedy algorithm is a method of local optimal selection. At each step, the optimal solution in the current state is selected and the path is gradually constructed. Greedy algorithm cannot guarantee to find the global optimal solution, but it can provide a good approximate solution in some cases.
[0087] Approximate Algorithms: Approximate algorithms are a class of algorithms that provide approximate solutions to a problem in polynomial time. For example, minimum spanning tree algorithms such as Primm’s algorithm and Kruskal’s algorithm can be used to approximate the solution of TSP.
[0088] Genetic algorithm: Genetic algorithm is a heuristic algorithm that searches the solution space by simulating natural selection and genetic mechanisms. It can be used to solve TSP, especially when dealing with large-scale problems.
[0089] Ant Colony Algorithm: Ant Colony Algorithm simulates the behavior of ants in the process of searching for food. Ant Colony Algorithm is widely used in TSP. Ants update the path through the deposition and evaporation of pheromones in the process of searching the path.
[0090] These methods have their own advantages and disadvantages and are suitable for TSP problems of different scales and characteristics. For small-scale problems, exhaustive methods and dynamic programming methods may be feasible; for large-scale problems, heuristic methods such as greedy algorithms, approximate algorithms, genetic algorithms, and ant colony algorithms may be more practical. Choosing a suitable method depends on the scale and characteristics of the problem and the accuracy requirements of the solution. Our application scenario requires fast calculation speed and simple algorithm implementation. Although the greedy algorithm cannot guarantee to find the optimal solution on TSP, in some cases, it can provide a good approximate solution with lower time complexity, so the greedy algorithm is selected for solution.
[0091] After sorting multiple key inspection areas, it is necessary to plan the path between two adjacent key inspection areas. Path planning refers to the process of finding the best path from the starting point to the target point in a given environment. This problem covers many different fields, including robot navigation, autonomous driving, logistics management, etc. The following are some common path planning methods:
[0092] Shortest path algorithm: Dijkstra algorithm finds the shortest path from the starting point to the target point through the shortest path between the nodes of the graph; A* algorithm combines Dijkstra algorithm and heuristic search to speed up the search process by estimating the distance to the target.
[0093] Graph-based search algorithms: Depth-first search (DFS) traverses the depth of the graph until the target is found or the maximum depth is reached; breadth-first search (BFS) expands the search layer by layer according to distance, ensuring that nodes close to the starting point are explored first.
[0094] Sampling-based search algorithms: Rapidly-Exploring Random Trees (RRT) generates paths by randomly sampling points in the configuration space and using rapid exploration to expand the tree structure; Probabilistic Roadmap (PRM) samples a large number of random points in the configuration space, builds a network, and then uses a graph search algorithm to find the path.
[0095] Optimization-based algorithms: Genetic algorithms optimize the quality of paths by simulating natural selection and genetic mechanisms; simulated annealing algorithms gradually reduce the system temperature by simulating the process of metal smelting to find the global optimal solution.
[0096] Artificial neural network: Deep reinforcement learning uses neural networks and reinforcement learning techniques to allow the system to find the best path through trial and error learning; model predictive control predicts future environmental conditions and plans the path by optimizing control inputs.
[0097] Hierarchical planning is to decompose the planning task into multiple levels to facilitate the handling of large-scale and complex environments.
[0098] Hybrid methods combine different methods to overcome the limitations of each method and improve the performance of path planning. The choice of path planning method usually depends on the nature of the specific problem, including the complexity of the environment, path constraints, real-time requirements, etc. In practical applications, it is often necessary to combine multiple methods to better adapt to different scenarios and needs.
[0099] At present, the conventional obstacle avoidance control faces the integrator model, which is often used to deal with the problem of complete constraints in the control system. It is not suitable for the non-complete characteristics of the wall-climbing robot, so we need to design a trackable path planning algorithm that satisfies the non-complete constraints. The typical method that can solve this problem now is the hybrid A* algorithm.
[0100] Classic A* only considers the two dimensions x and y, and hybrid A* introduces the θ dimension on this basis. In classic A*, the car is equivalent to being treated as a point mass, while hybrid A* adds θ to represent the orientation, and uses x and y to represent the center of the rear axle of the vehicle.
[0101] In this way, the search space changes from the previous two-dimensional table to the xy-θ state space.
[0102] In hybrid A*, it can be understood as expanding outward from the initial small line segment on a chessboard, each time a small curve can be expanded, and the node can be anywhere on the chessboard until it connects to the small line segment at the end point.
[0103] In the classic A* algorithm, each state grid is treated as a pixel point, that is, the robot path is composed of lines connecting the centers of each grid. The biggest difference between the Field D* algorithm and the classic A* is that it allows the path to end on the four sides of the grid, but the paths are also connected by straight line segments. The Hybrid A* algorithm allows the path to end inside the state grid, and the lines are replaced by some kind of curves, which conforms to the non-holonomic constraints.
[0104] In classic A*, the way to expand a node is to find adjacent nodes. However, since hybrid A* needs to consider vehicle dynamics and adds the θ dimension for this purpose, hybrid A* cannot simply assume that adjacent nodes are expansion targets. Its expansion needs to consider vehicle dynamics.
[0105] Another feature of hybrid A* is the introduction of the Reeds-Shepp curve generation mechanism. The Reeds-Shepp curve can connect the initial and final poses in a way that basically meets the vehicle dynamics constraints, but does not consider obstacle avoidance factors at all. Another commonly used curve is the Dubins curve. The RS curve is similar to the Dubins curve, both of which are paths of a circle plus a straight line. The difference is that the RS curve allows the vehicle to reverse.
[0106] The Reeds-Shepp curve consists of only three parts: a circular arc to the left, a circular arc to the right, and a straight line. In the incomplete model, this corresponds to turning left, turning right, and going straight. Therefore, the path is a series of circular arcs and straight lines constrained by the turning radius of the incomplete model.
[0107] Although the Reeds-Shepp curve does not consider obstacle avoidance, it is generated quickly enough to generate a path that conforms to dynamics and then verify whether the path meets the collision condition. Specifically, during the execution of hybrid A*, the Reeds-Shepp curve can be intermittently triggered to generate a curve connecting the current pose and the end pose, and then determine whether the collision condition is met. If it is met, the problem is solved. If not, the result is discarded and the normal search logic is continued.
[0108] This method of adding Reeds-Shepp curve to the traditional A* search greatly improves the search speed of hybrid A*. Because the terminal pose can also meet the requirements, it is often used in path planning of incomplete models.
[0109] However, in the present invention, the hybrid A* algorithm is improved because the non-holonomic constraint model in the technical field, that is, the wall-climbing robot, not only has non-holonomic constraints, but also has some other motion constraints and posture constraints.
[0110] The motion constraint is caused by the mechanical structure of the wall-climbing robot: the body of the wall-climbing robot is connected to many cables, which are located below the body of the wall-climbing robot. When the wall-climbing robot operates autonomously, if the head of the robot moves downward, the cables will have a certain probability of hindering or even interfering with the operation of the wall-climbing robot. Therefore, the mechanical structure prevents the head of the wall-climbing robot from moving downward during operation.
[0111] The posture constraint is because when the wall-climbing robot moves in a direction perpendicular or nearly perpendicular to the water-cooled wall tubes, the irregular and uneven boiler water-cooled wall will cause the wall-climbing robot to bump, causing damage to its wheels and the balancer loaded inside. Therefore, its posture must be constrained to prevent the wall-climbing robot from moving in a direction perpendicular to the water-cooled wall tubes. In other words, the angle (attitude angle) between the head of the wall-climbing robot and the vertical direction is in the range of [-45°, 45°].
[0112] According to the above constraints, the path planning algorithm of the wall-climbing robot with this type of constraints is improved, that is, the hybrid A* algorithm is improved, new methods are proposed in the two aspects of neighbor node selection and path trajectory fitting, and case experiments and analysis are carried out.
[0113] The idea of the greedy algorithm in solving the Traveling Salesman Problem (TSP) is to gradually build a path based on the local optimal choice until the end condition of the problem is met. TSP is a classic combinatorial optimization problem. The goal is to find a path that allows the traveling salesman to pass through each city once and only once, and finally return to the starting city, and minimize the total length of the path.
[0114] The following is a basic idea of using a greedy algorithm to solve TSP:
[0115] 1. Select a starting city: Select a city as the starting point.
[0116] 2. Greedy selection: In the current state, select the city that is closest to the current city and has not been visited as the next city to visit.
[0117] 3. Update status: Add the selected city to the route and mark it as visited.
[0118] 4. Repeated greedy selection: Repeat the above process until all cities have been visited.
[0119] 5. Return to the starting city: Connect the last city of the path to the starting city to form a closed path.
[0120] 6. End condition: The end condition of the problem is met, such as all cities have been visited.
[0121] 7. Optimize the path: After the end, you may need to perform some optimization on the path to ensure that the resulting path is the shortest path.
[0122] The idea of this greedy algorithm is to gradually build a path by selecting the optimal solution in the current state at each step.
[0123] The improved hybrid A* algorithm is a heuristic search algorithm used to find the shortest path from the starting point to the end point. It evaluates the priority of each node based on the evaluation function (heuristic function) and selects the node with the highest priority for expansion until the end point is found or all reachable nodes are expanded.
[0124] The steps to improve the hybrid A* algorithm to generate a path are as follows:
[0125] 1. Initialize the open list and the closed list.
[0126] The open list is used to store nodes to be explored, and the closed list is used to store nodes that have been explored.
[0127] 2. Add the starting node to the open list and set it as the current node to determine whether the open list is empty:
[0128] a. When the open list is not empty, perform the following steps:
[0129] (1) Select the node with the lowest f value (f = g + h, where g is the actual cost from the starting node to the current node and h is the estimated cost from the current node to the target node) from the open list as the current node.
[0130] (2) Determine whether the current node is the target node:
[0131] (a) If the current node is the target node, a path is generated according to the path approximation principle, and the path generation is completed;
[0132] (b) If the current node is not the target node, according to the constraint of the body attitude angle θ, traverse all neighboring nodes that meet the constraint and determine whether the path between the current node and the neighboring node passes through an obstacle:
[0133] (a) If an obstacle is passed, delete the neighbor node;
[0134] (b) If no obstacle is passed, determine whether it is in the closed list:
[0135] (a) If it is in the closed list, delete the node;
[0136] (b) If it is not in the closed list, calculate the g value of the neighbor node and determine whether the neighbor node has ever been in the open list:
[0137] (a) If it is in the open list, recalculate the g value of the neighbor node and compare the current g value of the neighbor node with the historical g value:
[0138] (a) If the current g value is less than the historical g value, update the neighbor node’s g = current g value;
[0139] (b) If the current g value ≥ the historical g value, keep the historical g value
[0140] (b) If it is not in the open list, calculate the h value and f value of the neighbor node, add the neighbor node to the open list, and set the parent node of the neighbor node as the current node.
[0141] b. If the open list is empty, it means the path cannot be found.
[0142] When the path is generated, you can trace back from the target node to the start node by backtracking the parent node to get the final path. In this process, the evaluation function is used to evaluate the distance from each node to the end point, and to select the node with the highest priority for expansion.
[0143] Figure 2 This is a case implementation process diagram based on the greedy algorithm and the improved hybrid A* algorithm. The algorithm generates the best detection order according to the greedy algorithm based on the environmental information, wall-climbing robot information, information on the area to be detected and obstacle information (inaccessible area information) given by the system, and arranges the detection order for a series of areas to be detected. Then, the improved hybrid A* algorithm is used to plan the path between two adjacent areas to be detected to generate a trackable, safe path with the lowest cost.
[0144] The task flow of the wall-climbing robot is as follows: the collaborative system sends the motion trajectory of the wall-climbing robot generated by the algorithm to the wall-climbing robot, and the wall-climbing robot starts from the hovering point and crawls to the key detection area (target climbing point, hereinafter collectively referred to as the target detection area). After reaching the key detection area, it performs dust cleaning, thickness measurement and other operations in the target detection area. In this process, the wall-climbing robot needs to feedback the current position, posture, image, speed, voltage and current and other information to the collaborative system in real time, and the collaborative system determines whether it is necessary to change the operating mode (manual / automatic) or whether it is necessary to return. If the operating mode needs to be changed or the return needs to be made, the wall-climbing robot is controlled to operate or return according to the new operating mode. If the operating mode does not need to be changed or the return needs to be made, the wall-climbing robot is controlled to crawl to the next key detection area and start a new round of key detection area operations.
[0145] Figure 3 This is a schematic diagram of the greedy algorithm solving the sorting problem of multiple areas to be detected. The diagram shows that the greedy algorithm can give a better traversal order to a certain extent, and the effect is good.
[0146] Figure 4 This is a simulation diagram of the improved hybrid A* algorithm for path planning of a wall-climbing robot. The gray box area in the figure is the inaccessible area in the environmental map, the stars mark the starting and ending positions of the wall-climbing robot's path, the dotted line is the ordinary path planning result generated by the A* algorithm, and the solid line is the optimized path planning result obtained by the improved hybrid A* algorithm.
[0147] Based on the classic A* algorithm, this algorithm introduces the motion constraints brought by the non-holonomic characteristics of the wall-climbing robot, the posture constraints in the actual operation process, and the position constraints brought by the inaccessible areas in the environment. Multiple constraints are considered together, and a curved path is used instead of the original broken line path to expand the classic A* algorithm. The purpose is to generate a shortest safe path that satisfies the kinematic constraints and can be tracked by the wall-climbing robot. The main innovations are as follows:
[0148] 1. The planning results are guaranteed to meet the kinematic constraints brought about by the non-holonomic characteristics of the wall-climbing robot, and to ensure the traceability of the planning results by the wall-climbing robot;
[0149] 2. Avoid overlapping of the output path in the planning results with the inaccessible areas in the environmental map to ensure the safety of the wall-climbing robot during operation.
[0150] The present invention performs path planning on the wall-climbing robot, and can provide the wall-climbing robot with a shortest safe path that satisfies kinematic constraints and can be tracked by the wall-climbing robot. At the same time, the rasterization processing of the environmental map can improve the precision and accuracy of the path planning of the wall-climbing robot. The wall-climbing robot operates according to the planning results, which can reduce the probability of injuries and other incidents caused by coking, manhole door slips, crashes, etc. during the movement of the wall-climbing robot, effectively reduce personnel safety accidents, improve system safety and reliability, ensure safe operation of the system, ensure safe production, and improve industrial economic benefits.
[0151] In a second aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm as described in any one of the first aspects above.
[0152] In a third aspect, an embodiment of the present invention provides a storage device, comprising a storage medium and a processor, wherein the storage medium stores a computer program, and when the program is executed by the processor, the robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm as described in any one of the first aspects above is implemented.
[0153] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0154] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include these modifications and variations.
[0155] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may alter, modify, replace and modify the above embodiments within the scope of the present invention.
Claims
1. A robot motion planning method based on a greedy algorithm and an improved hybrid A* algorithm, characterized in that: include: A greedy algorithm is used to determine the traversal order of the wall-climbing robot to traverse each detection area, which specifically includes: Select a detection area as a starting point; Select the detection area that is closest to the current detection area and has not been traversed as the next detection area; The next selected detection area is added to the path and marked as visited. The detection area closest to the current detection area and not traversed is selected as the next detection area until all detection areas are marked as visited, and the traversal order is determined after returning to the starting point. According to the traversal order, an improved hybrid A* algorithm is used to plan paths between adjacent detection areas.
2. The robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm according to claim 1, characterized in that: The method of using the improved hybrid A* algorithm to plan paths between adjacent detection areas according to the traversal order includes: Initialize the open list and the closed list, where the open list is used to store the nodes to be traversed, and the closed list is used to store the nodes that have been traversed; Add the starting point to the open list and set it as the current node, and determine whether the open list is empty; when the open list is not empty, select the node with the lowest f value from the open list as the current node, and plan the path based on whether the current node is the target node; when the open list is empty, no path can be found; where f = g + h, g is the actual cost from the starting point to the current node, and h is the estimated cost from the current node to the target node.
3. The robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm according to claim 2, characterized in that: The path planning according to whether the current node is a target node includes: If the current node is the target node, a path is generated according to the path approximation principle, and the path generation is completed; If the current node is not the target node, according to the constraint of the body attitude angle θ, traverse all neighboring nodes that meet the constraint, and perform path planning based on whether the path between the current node and the neighboring node passes through obstacles.
4. The robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm according to claim 1, characterized in that: The path planning according to whether the path between the current node and the neighboring node passes through an obstacle includes: If the path between the current node and the neighboring node passes through an obstacle, the neighboring node will be deleted; If the path between the current node and the neighboring node does not pass through obstacles, the path is planned based on whether the neighboring node is in the closed list.
5. The robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm according to claim 1, characterized in that: The path planning according to whether the neighbor node is in the closed list includes: If the neighbor node is in the closed list, delete the node; If the neighbor node is not in the closed list, calculate the g value of the neighbor node to determine whether the neighbor node has ever been in the open list; If the neighbor node was once in the open list, recalculate the g value of the neighbor node and determine the g value of the neighbor node based on the current g value and historical g value of the neighbor node; If the neighbor node is not in the open list, calculate the h value and f value of the neighbor node, add the neighbor node to the open list, and set the parent node of the neighbor node as the current node.
6. The robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm according to claim 1, characterized in that: Determining the g value of the neighbor node according to the current g value and the historical g value of the neighbor node includes: If the current g value of the neighbor node is less than the historical g value, update the g of the neighbor node to be equal to the current g value; If the current g value of the neighbor node is greater than or equal to the historical g value, keep the historical g value.
7. The robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm according to claim 1, characterized in that: The method further comprises: Control the robot to operate in each inspection area according to the planned path.
8. The robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm according to claim 1, characterized in that: The method further comprises: During the robot operation, the robot position, posture, image, speed, voltage and current information are fed back in real time; Determine whether to change the operation mode or return based on the robot's position, posture, image, speed, and voltage and current information.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm as described in any one of claims 1 to 8 is implemented.
10. A storage device comprising a storage medium and a processor, wherein the storage medium stores a computer program, wherein: When the processor executes the computer program, the robot motion planning method based on the greedy algorithm and the improved hybrid A* algorithm as described in any one of claims 1 to 8 is implemented.
Citation Information
Patent Citations
Method and device for planning target area operation, storage medium and processor
CN108846325A
Full coverage path planning method of sweeper
CN109459052A
Traversal planning method based on soft-wing unmanned aerial vehicle
CN110411445A
Autonomous coverage inspection method of intelligent robot in indoor complex dynamic environment
CN113110457A
Cattle farming type full-coverage path planning method and device based on synthesis unit
CN115014362A
Cited By
Path exploration parameter optimization system and method based on deep reinforcement learning
CN119828479A
A path exploration parameter optimization system and method based on deep reinforcement learning
CN119828479B
Game display control method and device, storage medium and electronic equipment
CN120617950A
