A path planning method for a traction substation inspection robot facing multiple task points

Abstract

The application relates to a traction substation inspection robot path planning method facing multiple task points, comprising the following steps: modeling the environment of the traction substation by using a grid map method to obtain an environment grid map, and calibrating task grid coordinates of multiple inspection task points on the environment grid map; based on the task grid coordinates, an improved swarm intelligence optimization algorithm is used to plan paths for the multiple inspection task points to obtain an optimal inspection sequence for accessing the multiple inspection task points; an improved real-time heuristic search algorithm is used to plan collision-free feasible segmented paths for each two adjacent inspection task points in the optimal inspection sequence, and all the feasible segmented paths are connected into a complete global inspection path; during movement of the inspection robot along the global inspection path, the real-time heuristic search algorithm is used to continuously optimize the unexecuted path segments in response to environmental changes and trigger path re-planning. The application can improve the inspection efficiency and real-time performance.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a path planning method for a traction substation inspection robot facing multiple task points. Background Technology

[0002] With the development of electrified railways and urban rail transit in my country, the traction power supply system, as the power source for train operation, is increasingly important for its safe, stable, and reliable operation. Traditional manual inspection methods have inherent limitations, including high labor intensity, long inspection cycles, susceptibility to subjective factors in testing standards, and high personal safety risks in high-risk environments. To achieve intelligent and unmanned operation and maintenance of traction substations, the construction of "unmanned substations" has become a clear industry trend. Against this backdrop, intelligent inspection robots with autonomous navigation and multi-sensor fusion capabilities are gradually replacing or assisting manual labor, undertaking daily inspection and monitoring tasks such as automatic equipment status identification, instrument reading, infrared temperature measurement, and anomaly alarms, becoming key technological equipment for ensuring the efficient operation of unmanned substations.

[0003] In recent years, numerous studies have introduced various intelligent algorithms into the field of robot path planning technology, attempting to solve problems such as environmental modeling, global optimization, and dynamic obstacle avoidance. However, for the scenario of traction substations, traditional technologies still face significant challenges: on the one hand, general-purpose global task sequence optimization algorithms are difficult to integrate into the special spatial and operational constraints of substations, such as dense equipment, regular passageways, and safety priorities; on the other hand, single refined path planning algorithms lack overall coordination of multi-task efficiency, while simple algorithm combinations suffer from problems such as inter-layer fragmentation and insufficient dynamic response. This makes it difficult for inspection robots to balance efficiency, optimality, and robustness in path planning, failing to fully meet the high standards of inspection efficiency and real-time performance required by unmanned substations. Summary of the Invention

[0004] Therefore, it is necessary to provide a path planning method for traction substation inspection robots that can improve inspection efficiency and real-time performance in the face of the above-mentioned technical problems.

[0005] In a first aspect, this application provides a path planning method for a traction substation inspection robot facing multiple task points, the method comprising:

[0006] The environment of the traction substation is modeled using the grid mapping method to obtain an environmental grid map, and the task grid coordinates of multiple inspection task points are marked on the environmental grid map.

[0007] Based on the task grid coordinates, an improved swarm intelligence optimization algorithm is used to perform path planning for the multiple inspection task points in order to obtain the optimal inspection sequence for visiting the multiple inspection task points.

[0008] An improved real-time heuristic search algorithm is used to plan a collision-free feasible segmented path for every two adjacent inspection task points in the optimal inspection sequence, and connect all feasible segmented paths into a complete global inspection path.

[0009] As the inspection robot moves along the global inspection path, the real-time heuristic search algorithm is used to continuously optimize unexecuted path segments and trigger path replanning in response to environmental changes.

[0010] In one embodiment, the step of performing path planning for the multiple inspection task points based on the task grid coordinates using an improved swarm intelligence optimization algorithm to obtain the optimal inspection sequence for visiting the multiple inspection task points includes:

[0011] For each particle iteration of the improved swarm intelligence optimization algorithm, the individual historical optimal solution for each particle in the current particle iteration is determined based on the current position information of each particle in the particle swarm and the individual historical optimal solution determined in the previous particle iteration. Based on the individual historical optimal solution of each particle in the current particle iteration, the target swarm historical optimal solution for the current particle iteration is determined. After each generation of particle updates, each particle represents a potential inspection sequence. Each potential inspection sequence contains multiple inspection task points with known task grid coordinates.

[0012] Based on the individual historical best solution of each particle in this particle iteration and the target group historical best solution of this particle iteration, update the current position information of each particle and enter the next particle iteration until the particle iteration stopping condition is met. The target group historical best solution determined in the last particle iteration is used as the optimal inspection sequence for visiting the multiple inspection task points.

[0013] In one embodiment, determining the target population historical optimal solution for the current particle iteration based on the individual historical optimal solution of each particle in this iteration includes:

[0014] Based on the individual historical best solution of each particle in this particle iteration, determine the initial swarm historical best solution for this particle iteration. ;

[0015] The initial population's historical optimal solution is considered as a ring, and two non-adjacent cut points are randomly selected. and Delete edge and Adding a new edge and and will and Reverse the subsequences between them to obtain the historical best solution for the target population. .

[0016] In one embodiment, determining the individual historical optimal solution for each particle in the current particle iteration based on the current position information of each particle in the particle swarm and the individual historical optimal solution determined in the previous particle iteration includes:

[0017] Based on the current position information of each particle in the particle swarm, determine the total theoretical movement distance for each particle;

[0018] The fitness of each particle in this iteration is determined by the reciprocal of the total theoretical movement distance for each particle.

[0019] The fitness of each particle in the current particle iteration is compared with the fitness of the individual historical best solution determined in the previous particle iteration, and the one with the higher fitness is taken as the individual historical best solution for each particle in the current particle iteration.

[0020] In one embodiment, updating the current position information of each particle based on its individual historical best solution for this particle iteration and the target group's historical best solution for this particle iteration includes:

[0021] ;

[0022] ;

[0023] in, Inertial weights are used to balance global and local search capabilities. and The acceleration constant, A random number within the interval [0,1]; Let be the velocity of the particle in the previous particle iteration. The velocity of the particle after the update. This represents the individual's historical optimal solution. This is the historically optimal solution for the target group. This provides the particle's current position information. This is the updated position information for the particles.

[0024] In one embodiment, the evaluation function of the improved real-time heuristic search algorithm is: ;

[0025] in, From the starting inspection point to the current inspection point of the current inspection path segment The actual cost, To start from the current inspection task point Heuristic cost estimation for reaching the target inspection task point on the current inspection path segment. As a heuristic weighting factor, and Greater than or equal to 1; during the continuous optimization of unexecuted path segments using the aforementioned real-time heuristic search algorithm, The iteration continues to decrease at a preset decreasing step size until the iteration stopping condition is met.

[0026] Secondly, this application provides a path planning device for a traction substation inspection robot facing multiple task points, the device comprising:

[0027] The module is used to model the environment of the traction substation using the grid mapping method, obtain an environmental grid map, and mark the task grid coordinates of multiple inspection task points on the environmental grid map.

[0028] The planning module is used to perform path planning for the multiple inspection task points based on the task grid coordinates using an improved swarm intelligence optimization algorithm to obtain the optimal inspection sequence for visiting the multiple inspection task points; it uses an improved real-time heuristic search algorithm to plan collision-free feasible segmented paths for every two adjacent inspection task points in the optimal inspection sequence, and connects all feasible segmented paths into a complete global inspection path; as the inspection robot moves along the global inspection path, it uses the real-time characteristics of the real-time heuristic search algorithm to continuously optimize unexecuted path segments and trigger path replanning in response to environmental changes.

[0029] Thirdly, this application provides an inspection robot, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the various method embodiments of this application.

[0030] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the various method embodiments of this application.

[0031] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps in the various method embodiments of this application.

[0032] The aforementioned path planning method for a traction substation inspection robot facing multiple task points employs a grid map method to model the environment of the traction substation, obtaining an environmental grid map, and defining the task grid coordinates of multiple inspection task points on the environmental grid map. Based on the task grid coordinates, an improved swarm intelligence optimization algorithm is used to plan paths for multiple inspection task points to obtain the optimal inspection sequence for visiting multiple inspection task points. An improved real-time heuristic search algorithm is used to plan collision-free feasible segmented paths for every two adjacent inspection task points in the optimal inspection sequence, and all feasible segmented paths are connected into a complete global inspection path. During the movement of the inspection robot along the global inspection path, the real-time heuristic search algorithm is used to continuously optimize unexecuted path segments and trigger path replanning in response to environmental changes. Compared to traditional technologies, this application uses a grid map to accurately model the unique environment of the substation. First, an improved swarm intelligence optimization algorithm is used to coordinate the overall task sequence, incorporating safety constraints such as equipment density and channel rules to obtain the optimal inspection order. Then, an improved real-time heuristic search algorithm is used to plan collision-free segmented paths and connect them into a complete path. Simultaneously, its real-time characteristics are utilized to continuously optimize unexecuted path segments and dynamically respond to environmental changes during the process. This overcomes the shortcomings of traditional solutions, such as inter-layer fragmentation and insufficient dynamic response, achieving a synergistic improvement in the overall efficiency, optimality, and dynamic adaptability of the inspection path, effectively meeting the high standards of inspection efficiency and real-time performance required by unattended substations. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a flowchart illustrating a path planning method for a traction substation inspection robot facing multiple task points in one embodiment.

[0035] Figure 2 This is a flowchart illustrating the process of controlling planning behavior using an improved real-time heuristic search algorithm in one embodiment.

[0036] Figure 3 This is a flowchart illustrating a path planning method for a traction substation inspection robot facing multiple task points, as described in another embodiment.

[0037] Figure 4 This example illustrates eight core inspection task points, tasks, and coordinate diagrams for each task point.

[0038] Figure 5 This is a schematic diagram of a blank environmental raster map containing random inspection task points in one embodiment;

[0039] Figure 6 This is a schematic diagram of the optimal inspection sequence under the improved swarm intelligence optimization algorithm in one embodiment;

[0040] Figure 7 This is a schematic diagram of the global inspection path in one embodiment;

[0041] Figure 8 This is a structural block diagram of a path planning device for a traction substation inspection robot facing multiple task points in one embodiment.

[0042] Figure 9 This is an internal structure diagram of an inspection robot in one embodiment. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0044] Currently, with the widespread application of mobile robots, their path planning technology has undergone several development stages, from simple pathfinding based on geometric models to real-time obstacle avoidance integrating environmental perception, and then to multi-objective decision-making incorporating intelligent optimization algorithms. In particular, the introduction and application of swarm intelligence optimization algorithms and real-time, incremental search algorithms have enabled robots to perform global and dynamic path planning in complex, unstructured environments, resulting in a qualitative improvement in autonomous navigation capabilities. However, in specific industrial scenarios such as traction substations with strict procedures, dense equipment, and high safety requirements, simply mechanically combining these traditional path planning algorithms reveals obvious limitations.

[0045] This application aims to overcome the limitations of simply stacking algorithms and proposes a deeply collaborative two-layer planning architecture. At the top layer, an improved swarm intelligence optimization algorithm (i.e., an improved PSO+2-opt algorithm) outputs an executable task sequence. At the bottom layer, an improved real-time heuristic search algorithm (i.e., an improved ARA* search algorithm) not only quickly finds paths but also feeds back the costs of obstacles encountered during execution to the top layer in real time, triggering re-optimization of the sequence. Through this closed loop of "planning-execution-feedback-replanning," the inspection robot's path can be calculated not only from the environmental grid map but also learned and continuously optimized from the actual work process and dynamic environment, thereby truly meeting the comprehensive requirements of unmanned substations for efficient, reliable, and standardized inspection tasks.

[0046] This application constructs a two-layer collaborative planning framework of "task sequence-fine path" and deeply integrates an improved swarm intelligence optimization algorithm (i.e., the improved PSO+2-opt algorithm) and an improved real-time heuristic search algorithm (i.e., the improved ARA* search algorithm) to achieve efficient multi-task point path planning for traction substation inspection robots. This application significantly improves the overall efficiency of inspection operations. By globally optimizing the inspection sequence, it avoids redundant movements caused by unreasonable visit sequences. It enhances real-time performance and reliability in complex environments. The "on-demand" nature of the improved ARA* search algorithm ensures that the inspection robot can obtain a feasible path and begin action in a very short time under any circumstances. The proposed layered architecture has a clear structure, and the improved algorithm parameters have explicit physical meanings, facilitating rapid configuration and adjustment by engineers according to the specific layout and maintenance requirements of different substations, thus enhancing the engineering practicality and scalability of the solution. It achieves autonomous and intelligent control of complex traction substation environments, realizing a synergistic multiplication of economic and social benefits.

[0047] In one embodiment, such as Figure 1 As shown, a path planning method for a traction substation inspection robot facing multiple task points is provided. This method can be applied to inspection robots and includes the following steps:

[0048] Step 102: The environment of the traction substation is modeled using the grid map method to obtain an environmental grid map, and the task grid coordinates of multiple inspection task points are marked on the environmental grid map.

[0049] Specifically, the inspection robot can use a grid mapping method to model the environment of the traction substation. The environment of the traction substation is divided into uniform grids to obtain an environmental grid map. The task grid coordinates of all inspection task points are then accurately marked on the obtained environmental grid map. In the environmental grid map, black grids represent impassable obstacles, and white grids represent areas where the inspection robot can freely traverse.

[0050] Step 104: Based on the task grid coordinates, an improved swarm intelligence optimization algorithm is used to perform path planning for multiple inspection task points in order to obtain the optimal inspection sequence for visiting multiple inspection task points.

[0051] In one embodiment, the inspection robot can use an improved swarm intelligence optimization algorithm (i.e., an improved PSO+2-opt algorithm) based on task grid coordinates to perform path planning for multiple inspection task points with the goal of minimizing the total theoretical travel distance of the inspection sequence, so as to obtain the optimal inspection sequence for visiting multiple inspection task points.

[0052] Step 106: An improved real-time heuristic search algorithm is used to plan a collision-free feasible segmented path for every two adjacent inspection task points in the optimal inspection sequence, and all feasible segmented paths are connected into a complete global inspection path.

[0053] Among them, an improved real-time heuristic search algorithm, namely the improved ARA* search algorithm, is adopted.

[0054] Step 108: During the process of the inspection robot moving along the global inspection path, the real-time heuristic search algorithm is used to continuously optimize the unexecuted path segments and trigger path replanning in response to environmental changes.

[0055] In the aforementioned path planning method for a traction substation inspection robot facing multiple task points, a grid map method is used to model the environment of the traction substation, obtaining an environmental grid map, and marking the task grid coordinates of multiple inspection task points on the environmental grid map. Based on the task grid coordinates, an improved swarm intelligence optimization algorithm is used to plan paths for multiple inspection task points to obtain the optimal inspection sequence for visiting multiple inspection task points. An improved real-time heuristic search algorithm is used to plan collision-free feasible segmented paths for every two adjacent inspection task points in the optimal inspection sequence, and connects all feasible segmented paths into a complete global inspection path. During the movement of the inspection robot along the global inspection path, the real-time characteristics of the real-time heuristic search algorithm are used to continuously optimize unexecuted path segments and trigger path replanning in response to environmental changes. Compared to traditional technologies, this application uses a grid map to accurately model the unique environment of the substation. First, an improved swarm intelligence optimization algorithm is used to coordinate the overall task sequence, incorporating safety constraints such as equipment density and channel rules to obtain the optimal inspection order. Then, an improved real-time heuristic search algorithm is used to plan collision-free segmented paths and connect them into a complete path. Simultaneously, its real-time characteristics are utilized to continuously optimize unexecuted path segments and dynamically respond to environmental changes during the process. This overcomes the shortcomings of traditional solutions, such as inter-layer fragmentation and insufficient dynamic response, achieving a synergistic improvement in the overall efficiency, optimality, and dynamic adaptability of the inspection path, effectively meeting the high standards of inspection efficiency and real-time performance required by unattended substations.

[0056] In one embodiment, based on task grid coordinates, an improved swarm intelligence optimization algorithm is used to plan paths for multiple inspection task points to obtain the optimal inspection sequence for visiting multiple inspection task points. This includes: for each particle iteration of the improved swarm intelligence optimization algorithm, determining the individual historical optimal solution for each particle in the current particle iteration based on the current position information of each particle in the particle swarm and the individual historical optimal solution determined in the previous particle iteration, and determining the target swarm historical optimal solution for the current particle iteration based on the individual historical optimal solution for each particle in the current particle iteration; wherein, after each generation of particles is updated, each particle represents a potential inspection sequence; each potential inspection sequence contains multiple inspection task points with known task grid coordinates; based on the individual historical optimal solution for each particle in the current particle iteration and the target swarm historical optimal solution for the current particle iteration, updating the current position information of each particle, and entering the next particle iteration, until the particle iteration stopping condition is met, and using the target swarm historical optimal solution determined in the last particle iteration as the optimal inspection sequence for visiting multiple inspection task points.

[0057] In the above embodiments, a particle swarm optimization mechanism is used to dynamically update the historical optimal solutions of individuals and the group in each iteration, and to continuously optimize the potential inspection sequence using location information until convergence to obtain the optimal inspection sequence. This method effectively integrates global search and local optimization capabilities, improves the solution accuracy and convergence efficiency of the optimal inspection sequence, and ensures the rapid generation of high-quality inspection task sequences under the complex constraints of traction substations.

[0058] In one embodiment, determining the target population historical optimal solution for the current particle iteration based on the individual historical optimal solution of each particle in this iteration includes: determining the initial population historical optimal solution for the current particle iteration based on the individual historical optimal solution of each particle in this iteration. The initial population's historical optimal solution is considered as a ring, and two non-adjacent cut points are randomly selected. and Delete edge and Adding a new edge and and will and Reverse the subsequences between them to obtain the historical best solution for the target population. .

[0059] Specifically, the inspection robot can determine the initial group historical optimal solution for the current particle iteration based on the individual historical optimal solution of each particle in this iteration. The initial group's historical optimal solution is considered as a ring containing an edge returning to the starting point. Two non-adjacent cut points are randomly selected. and The condition is satisfied that 1 ≤ i < j-1 ≤ N-1 (where N represents N inspection task points). The partial structure of the original path near the cutting point and its distance are:

[0060] .

[0061] Delete edge and Adding a new edge and and will and Reverse the subsequences between them to obtain the historical best solution for the target population. The partial structure and distance of the new path become:

[0062] .

[0063] In the above embodiments, by performing a 2-opt local search on the current population's historical optimal solution after each generation of particles is updated, the aim is to directly reduce the total path length by reconstructing the connection method of the two edges in the path. This can improve the local exploitation capability of PSO and prevent premature convergence.

[0064] In one embodiment, determining the individual historical optimal solution for each particle in the current particle iteration based on the current position information of each particle in the particle swarm and the individual historical optimal solution determined in the previous particle iteration includes: determining the total theoretical movement distance corresponding to each particle based on the current position information of each particle in the particle swarm; determining the fitness of each particle in the current particle iteration based on the reciprocal of the total theoretical movement distance corresponding to each particle; comparing the fitness of each particle in the current particle iteration with the fitness corresponding to the individual historical optimal solution determined in the previous particle iteration, and taking the one with the higher fitness as the individual historical optimal solution for each particle in the current particle iteration.

[0065] In one embodiment, the fitness of a particle can be calculated using a fitness function. The fitness function F(P) designed in this application... i The fitness function F(P) is the reciprocal of the total theoretical travel distance of the inspection sequence, with the goal of minimizing the total theoretical travel distance. i )for:

[0066]

[0067] Wherein, D(P) i ) is the inspection sequence The total theoretical travel distance is calculated based on the coordinates of the inspection task point on the environmental grid map. Assume the inspection task point p... k The coordinates are Then the total theoretical distance traveled is D(P) i )for:

[0068]

[0069] It is understandable that the total theoretical moving distance is the sum of Euclidean distances without considering obstacles, and it serves as the primary evaluation criterion for the quality of the inspection sequence, aiming to obtain the shortest global access sequence in a geometric sense.

[0070] In the above embodiments, the fitness is calculated by using the reciprocal of the total theoretical travel distance as the fitness value. The current fitness is compared with historical fitness values ​​to update the optimal historical solution for each individual. This mechanism intuitively maps path length to the optimization objective, guiding particles to converge towards shorter inspection sequences. This effectively improves the search efficiency and accuracy of the optimal solution, ensuring the rapid acquisition of high-quality inspection sequences in complex substation environments.

[0071] In one embodiment, the current position information of each particle is updated based on its individual historical best solution for this particle iteration and the target group's historical best solution for this particle iteration, including:

[0072] ;

[0073] ;

[0074] in, Inertial weights are used to balance global and local search capabilities. and The acceleration constant, A random number within the interval [0,1]; Let be the velocity of the particle in the previous particle iteration. The velocity of the particle after the update. This represents the individual's historical optimal solution. This is the historically optimal solution for the target group. This provides the particle's current position information. This is the updated position information for the particles.

[0075] In one embodiment, the evaluation function of the improved real-time heuristic search algorithm is: ;in, From the starting inspection point to the current inspection point of the current inspection path segment The actual cost, To start from the current inspection task point Heuristic cost estimation for reaching the target inspection task point on the current inspection path segment. As a heuristic weighting factor, and Greater than or equal to 1; during the continuous optimization of unexecuted path segments using a real-time heuristic search algorithm, The iteration continues to decrease at a preset decreasing step size until the iteration stopping condition is met.

[0076] In one embodiment, the inspection robot employs an improved real-time heuristic search algorithm (i.e., an improved ARA* search algorithm) to perform global inspection path search. The improved ARA* search algorithm utilizes heuristic weights... To control planning behavior, such as Figure 2 As shown, specifically: taking the generated optimal inspection sequence and the environmental grid map as input, a systematic segmented planning process transforms the theoretical access order of inspection task points into a physical path that the inspection robot can directly execute. In practical implementation, it first uses initial heuristic weighting factors... A fast ARA* search is performed on the first inspection task point pair in the optimal inspection sequence. The algorithm can output the initial feasible path for that segment in a very short time, ensuring that the inspection robot can start its work immediately. In the background, the system uses a decreasing step size... Continuously reduce heuristic weighting factors And by utilizing ARA*'s unique information reuse mechanism, the path for the same inspection segment is iteratively optimized until... The algorithm reduces the number of segments to 1 and obtains the theoretical shortest path or reaches a preset time limit. This process is applied sequentially to each adjacent pair of inspection task points in the optimal inspection sequence. Finally, all optimized segmented paths are concatenated in their global sequence order into a complete, continuous, and collision-free global inspection path that runs from the starting point through all task points. This method ensures that the planning process simultaneously satisfies the requirements of real-time execution, asymptotic optimality of the path, and high efficiency of computational resources.

[0077] In one embodiment, such as Figure 3 As shown, a grid map method is used to model the environment of the traction substation, obtaining an environmental grid map, and the task grid coordinates of multiple inspection task points are marked on the environmental grid map. Based on the task grid coordinates, an improved swarm intelligence optimization algorithm (i.e., the improved PSO+2-opt algorithm) is used to plan paths for multiple inspection task points to obtain the optimal inspection sequence for visiting multiple inspection task points. An improved real-time heuristic search algorithm (i.e., the improved ARA* search algorithm) is used to plan collision-free feasible segmented paths for every two adjacent inspection task points in the optimal inspection sequence, and all feasible segmented paths are connected into a complete global inspection path. During the movement of the inspection robot along the global inspection path, the real-time characteristics of the real-time heuristic search algorithm are used to continuously optimize unexecuted path segments and trigger path replanning in response to environmental changes.

[0078] For example, such as Figure 4As shown, eight core inspection task points, tasks, and the coordinates of each task point are listed. The distribution of each inspection task point on the blank environment raster map is as follows: Figure 5 As shown, Figure 6 This is a schematic diagram of the optimal inspection sequence under the improved swarm intelligence optimization algorithm. Figure 7 This is a schematic diagram of the final global inspection path.

[0079] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially, these steps are not necessarily executed in that order. Unless otherwise expressly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the above embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0080] Based on the same inventive concept, this application also provides a path planning device for a traction substation inspection robot facing multiple task points, for implementing the path planning method for a traction substation inspection robot facing multiple task points described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the path planning device for a traction substation inspection robot facing multiple task points provided below can be found in the limitations of the path planning method for a traction substation inspection robot facing multiple task points described above, and will not be repeated here.

[0081] In one embodiment, such as Figure 8 As shown, a path planning device 800 for a traction substation inspection robot facing multiple task points is provided. This device specifically includes:

[0082] Module 802 is used to model the environment of the traction substation using the grid map method, obtain an environmental grid map, and mark the task grid coordinates of multiple inspection task points on the environmental grid map.

[0083] The planning module 804 is used to perform path planning for multiple inspection task points based on task grid coordinates and an improved swarm intelligence optimization algorithm to obtain the optimal inspection sequence for visiting multiple inspection task points. An improved real-time heuristic search algorithm is used to plan collision-free feasible segmented paths for every two adjacent inspection task points in the optimal inspection sequence, and all feasible segmented paths are connected into a complete global inspection path. During the movement of the inspection robot along the global inspection path, the real-time heuristic search algorithm is used to continuously optimize unexecuted path segments and trigger path replanning in response to environmental changes.

[0084] In one embodiment, the planning module 804 is further configured to, for each particle iteration of the improved swarm intelligence optimization algorithm, determine the individual historical optimal solution of each particle in the current particle iteration based on the current position information of each particle in the particle swarm and the individual historical optimal solution determined in the previous particle iteration, and determine the target swarm historical optimal solution for the current particle iteration based on the individual historical optimal solution of each particle in the current particle iteration; wherein, after each generation of particles is updated, each particle represents a potential inspection sequence; each potential inspection sequence contains multiple inspection task points with known task grid coordinates; based on the individual historical optimal solution of each particle in the current particle iteration and the target swarm historical optimal solution for the current particle iteration, update the current position information of each particle, and enter the next particle iteration, until the particle iteration stopping condition is met, and use the target swarm historical optimal solution determined in the last particle iteration as the optimal inspection sequence for visiting multiple inspection task points.

[0085] In one embodiment, the planning module 804 is further configured to determine the initial population historical optimal solution for the current particle iteration based on the individual historical optimal solution of each particle in this particle iteration. The initial population's historical optimal solution is considered as a ring, and two non-adjacent cut points are randomly selected. and Delete edge and Adding a new edge and and will and Reverse the subsequences between them to obtain the historical best solution for the target population. .

[0086] In one embodiment, the planning module 804 is further configured to determine the total theoretical movement distance of each particle based on the current position information of each particle in the particle swarm; determine the fitness of each particle in the current particle iteration based on the reciprocal of the total theoretical movement distance of each particle; compare the fitness of each particle in the current particle iteration with the fitness of the individual historical best solution determined in the previous particle iteration, and take the one with the higher fitness as the individual historical best solution of each particle in the current particle iteration.

[0087] In one embodiment, the planning module 804 is also configured to execute the following formula:

[0088] ;

[0089] ;

[0090] in, Inertial weights are used to balance global and local search capabilities. and The acceleration constant, A random number within the interval [0,1]; Let be the velocity of the particle in the previous particle iteration. The velocity of the particle after the update. This represents the individual's historical optimal solution. This is the historically optimal solution for the target group. This provides the particle's current position information. This is the updated position information for the particles.

[0091] In one embodiment, the evaluation function of the improved real-time heuristic search algorithm is: ;in, From the starting inspection point to the current inspection point of the current inspection path segment The actual cost, To start from the current inspection task point Heuristic cost estimation for reaching the target inspection task point on the current inspection path segment. As a heuristic weighting factor, and Greater than or equal to 1; during the continuous optimization of unexecuted path segments using a real-time heuristic search algorithm, The iteration continues to decrease at a preset decreasing step size until the iteration stopping condition is met.

[0092] The aforementioned path planning device for a traction substation inspection robot facing multiple task points uses a grid map method to model the environment of the traction substation, obtaining an environmental grid map, and marking the task grid coordinates of multiple inspection task points on the environmental grid map. Based on the task grid coordinates, an improved swarm intelligence optimization algorithm is used to plan paths for multiple inspection task points to obtain the optimal inspection sequence for visiting multiple inspection task points. An improved real-time heuristic search algorithm is used to plan collision-free feasible segmented paths for every two adjacent inspection task points in the optimal inspection sequence, and connects all feasible segmented paths into a complete global inspection path. During the movement of the inspection robot along the global inspection path, the real-time heuristic search algorithm continuously optimizes unexecuted path segments and triggers path replanning in response to environmental changes. Compared to traditional technologies, this application uses a grid map to accurately model the unique environment of the substation. First, an improved swarm intelligence optimization algorithm is used to coordinate the overall task sequence, incorporating safety constraints such as equipment density and channel rules to obtain the optimal inspection order. Then, an improved real-time heuristic search algorithm is used to plan collision-free segmented paths and connect them into a complete path. Simultaneously, its real-time characteristics are utilized to continuously optimize unexecuted path segments and dynamically respond to environmental changes during the process. This overcomes the shortcomings of traditional solutions, such as inter-layer fragmentation and insufficient dynamic response, achieving a synergistic improvement in the overall efficiency, optimality, and dynamic adaptability of the inspection path, effectively meeting the high standards of inspection efficiency and real-time performance required by unattended substations.

[0093] The various modules in the path planning device for the traction substation inspection robot facing multiple task points described above can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the inspection robot in hardware form or independent of it, or they can be stored in the memory of the inspection robot in software form, so that the processor can call and execute the corresponding operations of each module.

[0094] In one embodiment, an inspection robot is provided, the internal structure of which can be shown in the following diagram. Figure 8As shown, the inspection robot includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides the environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a path planning method for a traction substation inspection robot facing multiple task points. The display unit of the inspection robot is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the inspection robot can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the shell of the inspection robot, or external keyboards, touchpads, or mice, etc.

[0095] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the inspection robot to which the present application is applied. A specific inspection robot may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0096] In one embodiment, an inspection robot is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0097] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0098] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0099] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0100] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0101] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0102] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A path planning method for a traction substation inspection robot facing multiple task points, characterized in that, The method includes: The environment of the traction substation is modeled using the grid mapping method to obtain an environmental grid map, and the task grid coordinates of multiple inspection task points are marked on the environmental grid map. Based on the task grid coordinates, an improved swarm intelligence optimization algorithm is used to perform path planning for the multiple inspection task points in order to obtain the optimal inspection sequence for visiting the multiple inspection task points. An improved real-time heuristic search algorithm is used to plan a collision-free feasible segmented path for every two adjacent inspection task points in the optimal inspection sequence, and connect all feasible segmented paths into a complete global inspection path. As the inspection robot moves along the global inspection path, the real-time heuristic search algorithm is used to continuously optimize unexecuted path segments and trigger path replanning in response to environmental changes.

2. The method according to claim 1, characterized in that, The step of using an improved swarm intelligence optimization algorithm to perform path planning on the multiple inspection task points based on the task grid coordinates to obtain the optimal inspection sequence for visiting the multiple inspection task points includes: For each particle iteration of the improved swarm intelligence optimization algorithm, the individual historical optimal solution for each particle in the current particle iteration is determined based on the current position information of each particle in the particle swarm and the individual historical optimal solution determined in the previous particle iteration. Based on the individual historical optimal solution of each particle in the current particle iteration, the target swarm historical optimal solution for the current particle iteration is determined. After each generation of particle updates, each particle represents a potential inspection sequence. Each potential inspection sequence contains multiple inspection task points with known task grid coordinates. Based on the individual historical best solution of each particle in this particle iteration and the target group historical best solution of this particle iteration, update the current position information of each particle and enter the next particle iteration until the particle iteration stopping condition is met. The target group historical best solution determined in the last particle iteration is used as the optimal inspection sequence for visiting the multiple inspection task points.

3. The method according to claim 2, characterized in that, The process of determining the target population's historical optimal solution for this particle iteration based on the individual historical optimal solution of each particle in this iteration includes: Based on the individual historical best solution of each particle in this particle iteration, determine the initial swarm historical best solution for this particle iteration. ; The initial population's historical optimal solution is considered as a ring, and two non-adjacent cut points are randomly selected. and Delete edge and Adding a new edge and and will and Reverse the subsequences between them to obtain the historical best solution for the target population. .

4. The method according to claim 2, characterized in that, The step of determining the individual historical optimal solution for each particle in the current particle iteration based on the current position information of each particle in the particle swarm and the individual historical optimal solution determined in the previous particle iteration includes: Based on the current position information of each particle in the particle swarm, determine the total theoretical movement distance for each particle; The fitness of each particle in this iteration is determined by the reciprocal of the total theoretical movement distance for each particle. The fitness of each particle in the current particle iteration is compared with the fitness of the individual historical best solution determined in the previous particle iteration, and the one with the higher fitness is taken as the individual historical best solution for each particle in the current particle iteration.

5. The method according to claim 2, characterized in that, The step of updating the current position information of each particle based on its individual historical best solution in this particle iteration and the target group's historical best solution in this particle iteration includes: ； ； in, Inertial weights are used to balance global and local search capabilities. and The acceleration constant, A random number within the interval [0,1]; Let be the velocity of the particle in the previous particle iteration. The velocity of the particle after the update. This represents the individual's historical optimal solution. This is the historically optimal solution for the target group. This provides the particle's current position information. This is the updated position information for the particles.

6. The method according to claim 1, characterized in that, The evaluation function of the improved real-time heuristic search algorithm is: ; in, From the starting inspection point to the current inspection point of the current inspection path segment The actual cost, To start from the current inspection task point Heuristic cost estimation for reaching the target inspection task point on the current inspection path segment. As a heuristic weighting factor, and Greater than or equal to 1; during the continuous optimization of unexecuted path segments using the aforementioned real-time heuristic search algorithm, The iteration continues to decrease at a preset decreasing step size until the iteration stopping condition is met.

7. A path planning device for a traction substation inspection robot facing multiple task points, characterized in that, The device includes: The module is used to model the environment of the traction substation using the grid mapping method, obtain an environmental grid map, and mark the task grid coordinates of multiple inspection task points on the environmental grid map. The planning module is used to perform path planning for the multiple inspection task points based on the task grid coordinates using an improved swarm intelligence optimization algorithm to obtain the optimal inspection sequence for visiting the multiple inspection task points; it uses an improved real-time heuristic search algorithm to plan collision-free feasible segmented paths for every two adjacent inspection task points in the optimal inspection sequence, and connects all feasible segmented paths into a complete global inspection path; as the inspection robot moves along the global inspection path, it uses the real-time characteristics of the real-time heuristic search algorithm to continuously optimize unexecuted path segments and trigger path replanning in response to environmental changes.

8. An inspection robot, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images