Nuclear power station inspection and maintenance robot path optimization method and system, and storage medium

Abstract

The invention discloses a nuclear power station inspection and maintenance robot path optimization method and system, and a storage medium, and the method comprises the steps: S1, carrying out the preliminary path planning through employing an APF algorithm, and obtaining a preliminary path; s2, performing global path search by using a Bi-RRT algorithm according to the initial path to obtain a global path; s3, according to the global path, performing local path optimization and smoothing processing by using an APF algorithm; and S4, taking the generated smooth and optimized path as a path for the robot to execute the task. By adopting the technical scheme of the invention, global path planning and local path optimization can be processed at the same time, the calculation efficiency is improved, and the smoothness and executable performance of the path are ensured.

Description

Technical Field

[0001] The present invention belongs to the field of information processing technology, and in particular relates to a path optimization method and system for a nuclear power plant inspection and maintenance robot, and a storage medium. Background Art

[0002] As a highly complex and potentially dangerous industrial environment, nuclear power plants place high demands on path planning. Inside a nuclear power plant, robots must not only avoid fixed obstacles such as equipment and walls, but also cope with dynamically changing environmental factors such as radiation sources, airflow, and temperature. The operating environment of a nuclear power plant is typically characterized by high temperatures, high radiation levels, confined spaces, and complex layouts, making the application of path planning algorithms even more challenging. During nuclear power plant inspection and maintenance tasks, robots must be able to efficiently and accurately plan paths, avoid dynamic obstacles, and ensure successful mission completion. This requires not only that path planning algorithms be able to effectively handle these changing environments, but also that the path planning results possess a high degree of stability and robustness. Therefore, a single path planning algorithm is often unable to cope with the complex and changing working environment of a nuclear power plant.

[0003] Although a variety of path planning algorithms have been proposed and applied in different industrial environments, the existing algorithms still have the following major problems in special and high-risk industrial environments such as nuclear power plants: ① Computational efficiency: Many traditional path planning algorithms have low computational efficiency in dynamic and complex environments such as nuclear power plants. In particular, when obstacles are dense or the environment changes, the calculation process may become very slow. ② Path quality: Existing path planning algorithms often generate paths that are not smooth enough or are long. In particular, the quality of the path will drop significantly under the influence of dynamic obstacles. ③ Local optimal solution problem: Although Bi-RRT can avoid local optimal solutions to a certain extent, it may still encounter problems of insufficiently smooth paths and local optimal solutions in complex environments, resulting in the robot being unable to effectively bypass obstacles. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a path optimization method and system for a nuclear power plant inspection and maintenance robot, as well as a storage medium, which can simultaneously process global path planning and local path optimization, improve computing efficiency, and ensure the smoothness and executability of the path.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for optimizing a nuclear power plant inspection and maintenance robot path, comprising:

[0007] Step S1: Use the APF algorithm to perform preliminary path planning to obtain a preliminary path;

[0008] Step S2: Based on the preliminary path, use the Bi-RRT algorithm to perform a global path search to obtain a global path;

[0009] Step S3: Based on the global path, use the APF algorithm to perform local path optimization and smoothing;

[0010] Step S4: Generate a smooth and optimized path as the path for the robot to perform the task.

[0011] Preferably, in step S1, the APF algorithm is used to guide the robot to move from the starting point toward the target point by calculating the attraction of the target and the repulsion of the obstacle, so as to obtain a preliminary path.

[0012] Preferably, in step S2, the Bi-RRT algorithm obtains the global path by expanding two trees from the starting point and the target point respectively and using a bidirectional expansion strategy.

[0013] The present invention also provides a nuclear power plant inspection and maintenance robot path optimization system, comprising:

[0014] The first processing module is used to perform preliminary path planning using the APF algorithm to obtain a preliminary path;

[0015] The second processing module is used to perform a global path search using the Bi-RRT algorithm based on the preliminary path to obtain a global path;

[0016] The third processing module is used to perform local path optimization and smoothing processing using the APF algorithm based on the global path;

[0017] The fourth processing module is used to generate a smooth and optimized path as the path for the robot to perform the task.

[0018] Preferably, the first processing module uses an APF algorithm to guide the robot to move from a starting point toward a target point by calculating the attraction of the target and the repulsion of the obstacle, thereby obtaining a preliminary path.

[0019] Preferably, the second processing module adopts the Bi-RRT algorithm to expand two trees from the starting point and the target point respectively, and obtains the global path using a bidirectional expansion strategy.

[0020] The present invention also provides a storage medium, on which a computer program is stored, and when the computer program is run, the method for optimizing the path of a nuclear power plant inspection and maintenance robot is executed.

[0021] The technical solution of the present invention, by combining global path planning with local path optimization, fully utilizes the global exploration capability of Bi-RRT and the local optimization advantages of APF, ensuring that high-quality and smooth paths can be quickly generated in complex and dynamic environments, meeting the requirements of nuclear power plant inspection and maintenance robots for efficient, accurate and robust path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0023] Figure 1 This is a flow chart of a path optimization method for a nuclear power plant inspection and maintenance robot according to an embodiment of the present invention;

[0024] Figure 2 Comparison diagram of the search paths of nuclear power plant inspection and maintenance robots in a two-dimensional environment; (a) is RRT, (b) is Bi-RRT, and (c) is Bi-RRT-APF;

[0025] Figure 3 Comparison diagram of the three-dimensional environment search path of the nuclear power plant inspection and maintenance robot; among them, (a) is RRT, (b) is Bi-RRT, and (c) is Bi-RRT-APF. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0027] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0028] Example 1:

[0029] like Figure 1 As shown, an embodiment of the present invention provides a path optimization method for a nuclear power plant inspection and maintenance robot, comprising:

[0030] Step S1: Use the APF algorithm to perform preliminary path planning to obtain a preliminary path;

[0031] Step S2: Based on the preliminary path, use the Bi-RRT algorithm to perform a global path search to obtain a global path;

[0032] Step S3: Based on the global path, use the APF algorithm to perform local path optimization and smoothing;

[0033] Step S4: Generate a smooth and optimized path as the path for the robot to perform the task.

[0034] Furthermore, in step S1, the present invention first uses the APF algorithm for preliminary path planning. During this process, the APF algorithm calculates the attractive force of the target and the repulsive force of obstacles to guide the robot from its starting point toward the target while avoiding collisions with obstacles. The APF algorithm can effectively avoid obstacles in a local area and provide the robot with a preliminary feasible path. However, because the APF algorithm is prone to local optimal solutions in complex environments, the present invention combines Bi-RRT to further optimize the path.

[0035] Furthermore, in step S2, to compensate for the APF algorithm's local optimal solution, the present invention uses the Bi-RRT algorithm to perform a global path search based on the initial path. By expanding two trees from the starting point and the target point, respectively, the Bi-RRT algorithm can effectively shorten the search path and improve path planning efficiency. Bi-RRT utilizes a bidirectional expansion strategy, avoiding the redundant extensions of the path search in the traditional RRT algorithm, and can find a feasible global path in a relatively short time.

[0036] Furthermore, in step S3, based on the global path generated by Bi-RRT, the present invention further uses the APF algorithm to optimize and smooth the local path. By calculating the gravitational force of the target and the repulsive force of obstacles, the APF algorithm can eliminate unnecessary bends and sharp turns in the path, ensuring the smoothness and feasibility of the path. Furthermore, the APF algorithm can effectively avoid local obstacles in the path, further improving the quality and stability of the path.

[0037] Furthermore, in step S4, the present invention connects and integrates the paths generated by the Bi-RRT and APF algorithms to form a complete path planning result. This step combines the advantages of both algorithms to ensure a globally optimal and locally smooth path. This optimized path not only effectively avoids obstacles but also ensures lower energy consumption, fewer sharp turns, and higher motion efficiency in actual execution.

[0038] The present invention combines the APF (Artificial Potential Field) and Bi-RRT (Bidirectionally Rapidly Expanding Random Tree) algorithms, combining the global path planning capabilities of Bi-RRT with the local path optimization capabilities of APF. Bi-RRT can simultaneously expand the path from both the starting point and the destination, reducing the search space and finding a path as quickly as possible, while APF optimizes the path, especially in terms of obstacle avoidance and local adjustments, preventing the robot from entering blind spots around obstacles and ensuring a smooth and executable path.

[0039] The Bi-RRT algorithm aims to find a path by extending two trees from the starting point and the target point, respectively. However, the Bi-RRT algorithm can produce a less-than-smooth path and may be disrupted by obstacles. The APF algorithm, on the other hand, locally optimizes each extended path. Especially in areas with dense obstacles or curved paths, APF uses the potential field to guide the robot to avoid obstacles and toward the target point. Combining these two methods can achieve both global planning and local obstacle avoidance, thereby improving path planning efficiency and path quality.

[0040] ① Initialize the bidirectional RRT tree

[0041] From the starting point x star and the target point x goal Initialize two trees T start and T goal , and set the maximum number of iterations max_iter, step size α and potential field parameters (gravitational coefficient k att and the repulsion coefficient k rep ).

[0042] ② Bidirectional RRT expansion

[0043] In each iteration, we start from the end nodes of the two trees. For each node, we first randomly generate a sampling point x rand , and then select the nearest node x according to the Bi-RRT rule near , and expand the tree according to the following formula:

[0044]

[0045] Among them, α is the expansion step size, which controls the growth rate of the tree.

[0046] ③ Calculation of combined force and APF optimization

[0047] During the expansion process, the artificial potential field method is used to calculate the resultant force at the current position. Each node is affected by the target attraction force and the obstacle repulsion force. The calculation formulas for the target attraction force and the obstacle repulsion force are:

[0048] F att (x) = -katt (xx goal )

[0049]

[0050] The robot extends the path in the direction of the resultant force. total (x) is the sum of the target attraction and the obstacle repulsion:

[0051] F total (x) = F att (x)+F rep (x)

[0052] The robot's position update formula is:

[0053]

[0054] This ensures that the robot moves towards the target point while avoiding obstacles as it expands.

[0055] ④Path connection

[0056] After each node is expanded, check whether the end nodes of the two trees meet. If so, connect the two trees and form a complete path:

[0057] ||x newstart -x newgoal ||<∈ (24)

[0058] where ∈ is the threshold at which the trees meet.

[0059] ④Path smoothing and optimization

[0060] The extended path is smoothed and optimized through APF, especially in terms of obstacle avoidance and local path adjustment. APF can effectively reduce path mutations, making the path smoother and more natural. The smoothed path can be optimized using the following formula:

[0061]

[0062] The optimized path will have smaller acceleration changes and higher stability, ensuring that the robot can move safely and smoothly in complex environments. If a path is found within the maximum number of iterations (max_iter), the path is returned; if not, a failure is returned.

[0063] The main configurations of nuclear power plant inspection and maintenance robots include the following: ① Wheeled configuration: suitable for flat and relatively simple ground environments, with good moving speed and flexibility; ② Tracked configuration: suitable for complex terrain, such as rugged, muddy or gravel ground, with strong passability and stability; ③ Wall-climbing configuration: using magnetic adsorption or vacuum adsorption technology, it can crawl on vertical or inclined surfaces such as walls and tanks of nuclear power plants, and is suitable for inspection and maintenance operations in these difficult-to-reach areas; ④ Articulated (robotic arm) configuration: with multiple degrees of freedom, it can simulate the movements of human arms and perform fine maintenance operations. It is usually carried on wheeled, tracked or other mobile platforms; ⑤ Underwater robot configuration: specially designed for underwater operations, such as the inspection and maintenance of underwater pipelines, valves and other equipment in nuclear power plants. According to the above configuration, the moving path of the nuclear power plant inspection and maintenance robot is divided into a two-dimensional environment and a three-dimensional environment. In order to verify the feasibility of the algorithm, MATLAB is used to define the starting point (0, 0) and the target point (100, 100) in the two-dimensional environment; the starting point (0, 0, 0) and the target point (100, 100, 100) are defined in the three-dimensional environment. The above four algorithms are used to perform path planning in the same environment. Each algorithm is repeated 50 times. The search path comparison diagram (at one time) is shown as follows: Figure 2-3 As shown, the step length (time) comparison is shown in Table 1.

[0064] Table 1

[0065] Method Name RRT Bi-RRT Bi-RRT-APF Average number of iterations 944 442 134 Average calculation time / s 27.12 3.73 1.90

[0066] contrast Figure 2 and Figure 3 It can be seen that the proposed Bi-RRT and APF fusion path optimization simulation algorithm for nuclear power plant inspection and maintenance robots is significantly superior to the traditional RRT and Bi-RRT algorithms in terms of average path length. Compared with the traditional RRT and Bi-RRT, the Bi-RRT and APF fusion algorithm reduces the average two-dimensional path length by approximately 25% and 15%, respectively, and the average three-dimensional path length by approximately 30% and 20%, respectively. In terms of path planning performance in the complex environment of nuclear power plant inspection and maintenance scenarios, the traditional RRT and Bi-RRT algorithms can effectively generate feasible paths in environments with fewer obstacles through a single expansion method, but Figure 2 (a) It can be seen that the number of trees extending outward from RRT is large and has no focus, resulting in poor path quality. Figure 3(a) It can be seen that 3D RRT is prone to falling into local optimal solutions, resulting in low search efficiency or even failure to find a path. Bi-RRT effectively speeds up the search process through bidirectional expansion, but it may still be affected by the distribution of obstacles, resulting in an insufficiently smooth or excessively long path. The fusion algorithm of Bi-RRT and APF fully combines the potential field advantages of APF guidance, effectively avoiding obstacles through the attraction and repulsion mechanism, and reducing the local optimal problem in the path generation process. The accuracy of the guidance force provided by APF in the target direction is combined with the efficiency of Bi-RRT bidirectional expansion. Figure 2 (c) with Figure 3 As shown in Figure (c), the fusion algorithm of Bi-RRT and APF ensures path smoothness. Particularly in complex or dynamic environments, APF's real-time adjustment capability makes the algorithm more robust and able to dynamically adapt to changes in obstacles. Furthermore, the fusion algorithm can better handle narrow passages and complex obstacle layouts, significantly improving the success rate and computational efficiency of path planning. Table 1 shows a comparison of path computation efficiency. Compared to traditional RRT and Bi-RRT, the fusion algorithm of Bi-RRT and APF significantly improves the average number of iterations and average computation time, with the average number of iterations decreasing by 85.8% and 69.7%, respectively, and the average computation time decreasing by 93.0% and 49.1%, respectively. Therefore, the fusion algorithm of Bi-RRT and APF has significant advantages in complex environments, particularly in improving path length and quality, computational efficiency, and avoiding local optimal solutions.

[0067] The embodiments of the present invention organically combine the global path planning capabilities of Bi-RRT with the local path optimization advantages of APF, providing an efficient and dynamic path planning solution for robots in complex environments. During the global path planning phase, Bi-RRT rapidly explores the path space through a bidirectional expansion tree approach, effectively reducing path search time, particularly in environments with dense obstacles or long paths. Simultaneously, the APF algorithm plays a key role in local path optimization, leveraging the attractive force of the target and the repulsive force of obstacles to guide the robot around obstacles and toward the target, thereby improving path quality. Specifically, APF optimizes the path by calculating the combined force, preventing the robot from entering blind spots or navigating unnecessary, redundant areas. Furthermore, APF's path smoothing function effectively reduces sharp turns and unbalanced motion within the path, ensuring smoother and more efficient robot execution. Overall, this fusion algorithm demonstrates high adaptability in the complex and dynamically changing nuclear power plant environment. It not only adjusts the path in real time based on environmental changes, but also ensures the robot can safely and reliably execute its tasks in complex and risky environments, significantly improving the overall efficiency and safety of path planning.

[0068] Example 2:

[0069] An embodiment of the present invention further provides a nuclear power plant inspection and maintenance robot path optimization system, comprising:

[0070] The first processing module is used to perform preliminary path planning using the APF algorithm to obtain a preliminary path;

[0071] The second processing module is used to perform a global path search using the Bi-RRT algorithm based on the preliminary path to obtain a global path;

[0072] The third processing module is used to perform local path optimization and smoothing processing using the APF algorithm based on the global path;

[0073] The fourth processing module is used to generate a smooth and optimized path as the path for the robot to perform the task.

[0074] As an implementation of an embodiment of the present invention, the first processing module uses the APF algorithm to guide the robot to move from the starting point toward the target point by calculating the attraction of the target and the repulsion of the obstacle, thereby obtaining a preliminary path.

[0075] As an implementation method of the embodiment of the present invention, the second processing module adopts the Bi-RRT algorithm to expand two trees from the starting point and the target point respectively, and uses a bidirectional expansion strategy to obtain a global path.

[0076] Example 3:

[0077] An embodiment of the present invention further provides a storage medium having a computer program stored thereon, and the computer program executes a path optimization method for a nuclear power plant inspection and maintenance robot when running.

[0078] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. A path optimization method for a nuclear power plant inspection and maintenance robot, characterized in that: include: Step S1: Use the APF algorithm to perform preliminary path planning to obtain a preliminary path; Step S2: Based on the preliminary path, use the Bi-RRT algorithm to perform a global path search to obtain a global path; Step S3: Based on the global path, use the APF algorithm to perform local path optimization and smoothing; Step S4: Generate a smooth and optimized path as the path for the robot to perform the task.

2. The nuclear power plant inspection and maintenance robot path optimization method according to claim 1, characterized in that: In step S1, the APF algorithm is used to guide the robot to move from the starting point to the target point by calculating the attraction of the target and the repulsion of the obstacles to obtain a preliminary path.

3. The nuclear power plant inspection and maintenance robot path optimization method according to claim 2, characterized in that: In step S2, the Bi-RRT algorithm obtains the global path by expanding two trees from the starting point and the target point respectively using a bidirectional expansion strategy.

4. A nuclear power plant inspection and maintenance robot path optimization system, characterized in that: include: The first processing module is used to perform preliminary path planning using the APF algorithm to obtain a preliminary path; The second processing module is used to perform a global path search using the Bi-RRT algorithm based on the preliminary path to obtain a global path; The third processing module is used to perform local path optimization and smoothing processing using the APF algorithm based on the global path; The fourth processing module is used to generate a smooth and optimized path as the path for the robot to perform the task.

5. The nuclear power plant inspection and maintenance robot path optimization system according to claim 4, characterized in that: The first processing module uses the APF algorithm to guide the robot to move from the starting point to the target point by calculating the attraction of the target and the repulsion of the obstacles to obtain a preliminary path.

6. The nuclear power plant inspection and maintenance robot path optimization system according to claim 5, characterized in that: The second processing module adopts Bi-RRT algorithm to expand two trees from the starting point and the target point respectively, and obtains the global path using the bidirectional expansion strategy.

7. A storage medium, characterized in that: The storage medium stores a computer program, which, when running, executes the path optimization method for a nuclear power plant inspection and maintenance robot according to any one of claims 1 to 3.

Images