Automatic inspection AGV dynamic path planning method, device and equipment and medium

Abstract

The invention discloses an autonomous inspection AGV dynamic path planning method, device and equipment and a medium, and the method comprises the steps: building a grid map based on environment information, and setting an initial position coordinate and an initial target position coordinate of an autonomous inspection AGV; initializing a plurality of parameters, and based on an RRT algorithm, a self-adaptive target offset probability strategy, improving heuristic search of an artificial potential field method and a variable self-adaptive step length function, obtaining a path track of the autonomous inspection AGV; optimizing the path trajectory based on a pruning optimization strategy of target backtracking, and extracting key nodes in the path trajectory after pruning optimization as sub-target points of local path planning; and under the guidance of the sub-target points, performing local dynamic path planning based on an artificial potential field method and a dynamic gravitational field strategy. The invention belongs to the field of automatic inspection. The method can be suitable for an unknown dynamic environment, the automatic inspection AGV can avoid dynamic obstacles, and tasks can be completed efficiently, stably and safely.

Description

Technical Field

[0001] The present invention relates to the field of automated inspection, and particularly to a method, device, equipment and medium for dynamic path planning of an autonomous inspection AGV. Background Art

[0002] In recent years, the inspection technology of unmanned autonomous vehicles has received increasing attention. It has many advantages, such as cost reduction, convenience, flexibility, safety, reliability, etc. In practical applications, since unmanned autonomous vehicles can obtain clear images and can reach many areas that are difficult for inspectors to access, they have been widely used in many fields, such as oilfield inspection, obstacle intrusion monitoring, disaster relief, healthcare, agricultural plant protection, etc. Nowadays, the development of the economy affects people's living standards, and the development of society poses higher challenges to the stable operation of the power system. Specifically, the inspection requirements for various links in the power system, such as substations, power transmission, and power distribution, have been further improved.

[0003] For example, in the power system inspection process, there are still a series of problems caused by the lack of timely on-site monitoring and patrol. There are many factors affecting the quality of electrical equipment detection, such as the planning of the detection route, detection tools, detection time arrangement, and other objective factors that may affect the detection quality.

[0004] So far, most power system autonomous inspection AGVs can be roughly divided into two types: rail-guided and rail-free. The former is more direct and reliable in terms of control and positioning, but its operation is non-autonomous and can only perform certain specific inspection tasks, and the inspection path is relatively fixed. The latter models the complex environment to achieve high-precision positioning and navigation. In practical applications, the power system autonomous inspection AGV can detect unknown obstacles and moving obstacles through corresponding built-in sensors when moving, so as to achieve autonomous navigation and dynamic path planning of the power system autonomous inspection AGV.

[0005] Therefore, the power system autonomous inspection AGV can not only construct a full-range map during the unmanned automated power system inspection process, but also effectively plan the critical path inspection of the complex path on-site. The intelligent mobile AGV for power system inspection should carry various instruments and sensors to replace manual labor to perform relevant power equipment inspections in an autonomous or remotely controlled manner. The power system autonomous inspection AGV can timely detect problems and defects such as whether the power equipment is damaged, diagnose potential hazards and precursors of accidents during the operation of the power equipment, so as to timely discover problems such as equipment defects, abnormalities, and intrusion of moving targets, and thus take measures in time to eliminate potential accident hazards. In the power system, the use of autonomous inspection AGVs can improve work efficiency and quality, reduce human resources, and lower production costs.

[0006] Based on this, the present invention provides a dynamic path planning method for autonomous inspection AGV, which is used to solve the path planning and dynamic obstacle avoidance problems of autonomous inspection AGV in the power system. Summary of the Invention

[0007] By providing a dynamic path planning method, device, equipment and medium for autonomous inspection AGV, the present invention solves the technical problem of difficult obstacle avoidance of autonomous inspection AGV in the prior art, and can achieve the technical effect of automatic obstacle avoidance of autonomous inspection AGV.

[0008] In the first aspect, the present invention provides a dynamic path planning method for autonomous inspection AGV, and the method includes: Establish a grid map based on environmental information, and set the initial position coordinates and initial target position coordinates of the autonomous inspection AGV; Initialize several parameters, and based on the RRT algorithm, adaptive target biasing probability strategy, heuristic search of the improved artificial potential field method, and variable adaptive step function, achieve global path planning to obtain the path trajectory of the autonomous inspection AGV; Based on the pruning optimization strategy of target backtracking, optimize the path trajectory, and extract the key nodes in the pruned and optimized path trajectory as sub-goals for local path planning; Under the guidance of the sub-goal points, perform local dynamic path planning based on the artificial potential field method and the dynamic gravitational field strategy.

[0009] Further, establishing a grid map based on environmental information and setting the initial position coordinates and initial target position coordinates of the autonomous inspection AGV includes: Preprocess the environmental information, including: outlier processing, missing value processing, and feature extraction; After preprocessing, establish a grid map according to the environmental information; In the grid map, set the initial position coordinates and initial target position coordinates of the autonomous inspection AGV, and create the inspection task content of the autonomous inspection AGV.

[0010] Further, initializing several parameters includes: Initialize the obstacle coordinates, initial position coordinates, and initial target position coordinates of the autonomous inspection AGV; Initialize the initial step length, target biasing probability value, safety factor of the biasing probability, gravitational gain coefficient and repulsive gain coefficient of the improved artificial potential field method of the RRT algorithm.

[0011] Further, based on the RRT algorithm, adaptive target biasing probability strategy, heuristic search of the improved artificial potential field method, and variable adaptive step function, achieving global path planning to obtain the path trajectory of the autonomous inspection AGV includes: Construct an adaptive target biasing probability strategy and determine the sampling position of the random sampling points of the RRT algorithm, including: , wherein, is an arbitrary number between 0 and 1, is the probability biasing threshold, is the sampling position of the random sampling point determined according to the adaptive probability biasing strategy, is the initialized target position coordinate point, is the current random sampling point obtained by the RRT algorithm; wherein, the probability biasing threshold , includes: , wherein, is the area occupied by obstacles within a circular region centered at the current coordinate of the autonomous inspection AGV and with a radius equal to the length of the current step of the autonomous inspection AGV, is the length of the current step of the autonomous inspection AGV, is the safety factor; Construct a heuristic search based on the improved artificial potential field method, and guide the RRT algorithm to generate random points through the improved artificial potential field method. Among them, the gravitational field and repulsive field functions of the improved artificial potential field method include: , , wherein, is the gravitational field gain coefficient, is the repulsive field gain coefficient, is the distance from the current node to the target position, is the distance from the current node to the center of action of the repulsive field, is the radius of action of the repulsive field, is a positive integer, is the repulsive field, is the gravitational field; The gravitational field and repulsive field functions include: , , , , wherein, is the unit vector pointing from the current node to the target point, is the unit vector pointing from the obstacle to the current node, is the gravitational value, is the repulsive value pointing from the current node to the target point, The value obtained by taking the partial derivative of the nth power of the distance between the current node and the target node, The repulsive force value directed from the obstacle to the current node, The total repulsive force value of the current node, The repulsive force field; Construct a variable adaptive step-size function to regulate the step size of the RRT algorithm. The variable adaptive step-size function includes: , where, is the step size updated based on the adaptive step-size function, is the gravitational force of the current node, is the gravitational force of the target position with respect to the initial position, is the step size of the parent node of the current node, is the distance between the current node and the center of action of the repulsive force field, is the repulsive force value of the current node.

[0012] Furthermore, based on the pruning optimization strategy of target backtracking, optimize the path trajectory, and extract the key nodes in the pruned and optimized path trajectory as the sub-goal points for local path planning, including: Step S131: Take the target point in the result of the global path planning as the current node, and perform path backtracking based on the target point; Step S132: Take the current node as the parent node and the previous node of the current node as the child node, and determine whether the line connecting the parent node and the child node collides with an obstacle; Step S133: If the line connecting the parent node and the child node does not collide with an obstacle, take this child node as a key node and keep the current node still as the parent node; Step S134: Take the previous node of the child node as the new child node, and determine whether the line connecting the new child node and the parent node collides with an obstacle; Step S135: If there is no collision, take the new child node as a key node, remove the old child node in Step S134, and repeat Steps S134 - S135; if there is a collision, take the old child node in Step S134 as the new parent node and re-execute Steps S132 - S135; After backtracking to the initial position coordinates, reconstruct the path planning scheme based on the key nodes and terminate the pruning optimization strategy, and extract the optimized key nodes as the sub-goal points for local path planning.

[0013] Furthermore, under the guidance of the sub-goal points, perform local dynamic path planning based on the artificial potential field method and combined with the dynamic gravitational field strategy, including: Arrange the sub-goal points in sequence to form a set of sub-goal points; In local path planning, sequentially extract the sub-goal points in the set of sub-goal points as the current goal point of the artificial potential field method; Among them, the dynamic gravitational field function is: , In the formula: is the gain coefficient; is the gravitational field gain coefficient; is the distance between the current node and the intersection point obtained by projecting the current node onto the global path, is the gravitational value of the current node.

[0014] Furthermore, perform missing value processing on the environmental information, including: Based on the linear interpolation method, process the missing values in the environmental information.

[0015] In a second aspect, the present invention provides an autonomous inspection AGV dynamic path planning device, which includes: A map construction module for establishing a grid map based on environmental information and setting the initial position coordinates and initial target position coordinates of the autonomous inspection AGV; An initialization module for initializing several parameters and implementing global path planning based on the RRT algorithm, adaptive target biasing probability strategy, heuristic search of the improved artificial potential field method, and variable adaptive step function to obtain the path trajectory of the autonomous inspection AGV; Based on the pruning optimization strategy of target backtracking, optimize the path trajectory and extract the key nodes in the pruned and optimized path trajectory as the sub-goals of local path planning; Under the guidance of the sub-goal points, perform local dynamic path planning based on the artificial potential field method and combined with the dynamic gravitational field strategy.

[0016] In a third aspect, the present invention provides an electronic device, including: A processor; A memory for storing instructions executable by the processor; Among them, the processor is configured to execute to implement an autonomous inspection AGV dynamic path planning method provided in the first aspect.

[0017] In a fourth aspect, the present invention provides a non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by the processor of the electronic device, enabling the electronic device to execute and implement an autonomous inspection AGV dynamic path planning method as provided in the first aspect.

[0018] One or more technical solutions provided in the present invention have at least the following technical effects or advantages: An adaptive target biasing probability strategy is constructed. During the growth process of the random tree, the environmental complexity around the parent node of the new node to be generated is judged, that is, the area occupied by obstacles, and then the specific biasing probability is further determined. At the same time, an adaptive step-size function is constructed. During the growth process of the random tree, the environmental complexity around the parent node of the new node to be generated, the number of algorithm iterations, and the relative distance between the random tree and the target point are judged, and then the value of the adaptive step-size function is further determined. Based on generating higher-quality new nodes during the expansion of the random tree, the algorithm can plan a higher-quality path with higher efficiency. At the same time, combined with a pruning optimization strategy based on backtracking the path from the target point, redundant points in the global path are removed, further improving the path quality.

[0019] A method for local path planning in a dynamic environment with global key nodes as local sub-target points is constructed, which solves the problems that the target is unreachable and it is easy to fall into local optimality when the autonomous inspection AGV performs local path planning through the artificial potential field method. Through the local path algorithm optimization strategy based on the dynamic gravitational field, the coordination mechanism problem of integrating the global and local planning results is solved, thereby effectively reducing the path length of the autonomous inspection AGV to reach the target point. Brief Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0021] Figure 1 It is a schematic flowchart of a dynamic path planning method for an autonomous inspection AGV provided by the present invention; Figure 2 It is a schematic diagram of heuristic search based on an improved artificial potential field method provided by the present invention; Figure 3 It is a schematic diagram of extracting key nodes in the pruned and optimized path as sub-target points for local path planning provided by the present invention; Figure 4 It is a schematic diagram of the dynamic gravitational field provided by the present invention; Figure 5 It is a schematic diagram of the path planning result provided by the present invention. Detailed Embodiments

[0022] By providing a dynamic path planning method for an autonomous inspection AGV in the embodiments of the present invention, the technical problem of difficult obstacle avoidance of the autonomous inspection AGV in the prior art is solved.

[0023] The technical solution of the present invention for solving the above technical problems has the following general idea: An autonomous inspection AGV dynamic path planning method, the method includes: establishing a grid map based on environmental information, and setting the initial position coordinates and initial target position coordinates of the autonomous inspection AGV; initializing a number of parameters, and based on the RRT algorithm, adaptive target biasing probability strategy, heuristic search of the improved artificial potential field method, and variable adaptive step function, realizing global path planning to obtain the path trajectory of the autonomous inspection AGV; based on the pruning optimization strategy of target backtracking, optimizing the path trajectory, and extracting the key nodes in the pruned and optimized path trajectory as sub-goals for local path planning; under the guidance of the sub-goal points, performing local dynamic path planning based on the artificial potential field method and the dynamic gravitational field strategy.

[0024] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the specification drawings and specific implementation manners.

[0025] First, it should be noted that the term "and / or" appearing in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0026] The present invention provides Figure 1 an autonomous inspection AGV dynamic path planning method as shown, including steps S11 - S14: Step S11, establishing a grid map based on environmental information, and setting the initial position coordinates and initial target position coordinates of the autonomous inspection AGV.

[0027] The autonomous inspection AGV (Automated Guided Vehicle) can collect environmental information through a variety of sensors equipped on it {such as lidar (LiDAR), camera, infrared sensor, ultrasonic sensor, etc.}, and then construct a grid map.

[0028] The grid map is a method of discretizing a continuous space, dividing the entire environment into small square or rectangular regions (i.e., grids), and each grid is assigned a certain attribute value (for example: passability, presence or absence of obstacles, etc.). The grid map helps to simplify the path planning algorithm and enables the autonomous inspection AGV to more effectively understand and process the surrounding environment.

[0029] Build a grid map based on environmental information, and set the initial position coordinates and initial target position coordinates of the autonomous inspection AGV, including: preprocess the environmental information, including outlier processing, missing value processing, and feature extraction; after preprocessing, build a grid map according to the environmental information; in the grid map, set the initial position coordinates and initial target position coordinates of the autonomous inspection AGV, and create the inspection task content of the autonomous inspection AGV.

[0030] After obtaining the sensor data, a series of data processing processes can be carried out, including data cleaning, feature extraction, and data fusion, etc., to extract the effective information required for building the grid map.

[0031] Perform missing value processing on the environmental information, including: based on the linear interpolation method, process the missing values in the environmental information.

[0032] In step S12, initialize several parameters, and based on the RRT algorithm, adaptive target bias probability strategy, heuristic search of the improved artificial potential field method, and variable adaptive step size function, implement global path planning to obtain the path trajectory of the autonomous inspection AGV; Initialize several parameters, including: initialize the obstacle coordinates, initial position coordinates, and initial target position coordinates of the autonomous inspection AGV; initialize the initial step length, target bias probability value, safety factor of the bias probability, gravitational gain coefficient and repulsive gain coefficient of the improved artificial potential field method of the RRT algorithm.

[0033] Based on the RRT algorithm, adaptive target bias probability strategy, heuristic search of the improved artificial potential field method, and variable adaptive step size function, implement global path planning to obtain the path trajectory of the autonomous inspection AGV, including: Construct an adaptive target bias probability strategy, and determine the sampling position of the random sampling point of the RRT algorithm, including: , Among them, is an arbitrary number between 0 and 1, is the probability bias threshold, is the sampling position of the random sampling point determined according to the adaptive probability bias strategy, is the initialized target position coordinate point (i.e., the initial target position coordinate), is the current random sampling point obtained by the (improved) RRT algorithm; among them, the growth direction of the random tree is determined by the probability bias threshold decides that when the random number is less than the probability bias threshold the sampling node growing in the random tree is a randomly generated node, when the random number When it is greater than or equal to the probability bias threshold, the sampling node of the random tree is the target point in the grid map.

[0034] During the growth process of the random tree, the probability of successfully generating a new node is usually determined by the environmental complexity around its parent node, that is, the area occupied by obstacles. Depending on the environment where the parent node is located, when the parent node is in an open environment or there are fewer obstacles around it, the probability of successfully generating a new node is significantly higher than when the parent node is in a complex environment with more obstacles around it. Therefore, the probability bias threshold , including: , Among them, is the area occupied by obstacles within the circular area with the current coordinates of the autonomous inspection AGV as the center and the length of the current step of the autonomous inspection AGV as the radius, is the length of the current step of the autonomous inspection AGV, is the safety factor; Construct a heuristic search based on the improved artificial potential field method, and use the improved artificial potential field method to guide the RRT algorithm to generate random points. Among them, the gravitational field and repulsive field functions of the improved artificial potential field method include: , , Among them, is the gravitational field gain coefficient, is the repulsive field gain coefficient, is the distance from the current node to the target position, is the distance from the current node to the center of the repulsive field action, is the repulsive field action radius, is a positive integer, is the repulsive field, is the gravitational field; The gravitational field and repulsive field functions include: , , , , Among them, is the unit vector from the current node to the target point, is the unit vector from the obstacle to the current node, is the gravitational value, is the repulsive value from the current node to the target point, is the value obtained by taking the partial derivative of the nth power of the distance between the current node and the target node, is the repulsive force value pointing from the obstacle to the current node, is the total repulsive force value of the current node, is the repulsive force field; Figure 2 is the schematic diagram of heuristic search based on the improved artificial potential field method provided by the present invention. Both the target point and the obstacle exert a repulsive force on the point. The repulsive force exerted by the target point on the point points from the point to the target point, and the repulsive force exerted by the obstacle on the point points from the obstacle to the point. and are the total resultant repulsive force synthesized according to the parallelogram law. Similarly, is synthesized by the component and . Based on the heuristic search of the improved artificial potential field method, when the point coordinates approach the target point, the repulsive force will also decrease accordingly, and the repulsive force becomes zero after reaching the target point. The position where the new node is generated is determined by the vector from the random point to the nearest node, the gravitational vector from the target point to the nearest node, and the resultant repulsive force vector of the obstacle on the nearest node. Finally, the new node is located in the direction of the vector.

[0035] Construct a variable adaptive step size function to regulate the step size of the RRT algorithm. The variable adaptive step size function includes: , where is the step size updated based on the adaptive step size function, is the gravitational force of the current node, is the gravitational force of the target position on the initial position, is the step size of the parent node of the current node, is the distance between the current node and the center of the repulsive force field, is the repulsive force value of the current node.

[0036] Step S13, based on the pruning optimization strategy of target backtracking, optimize the path trajectory, and extract the key nodes in the pruned and optimized path trajectory as the sub-goals of local path planning; Specifically, it includes steps S131 - S135: Step S131, take the target point in the result of global path planning as the current node, and backtrack the path based on the target point; Step S132: Taking the current node as the parent node and the previous node of the current node as the child node, determine whether the connection line between the parent node and the child node collides with an obstacle; Step S133: If there is no collision between the connection line between the parent node and the child node and an obstacle, take the child node as the key node and keep the current node still as the parent node; Step S134: Take the previous node of the child node as the new child node and determine whether the connection line between the new child node and the parent node collides with an obstacle; Step S135: If there is no collision, take the new child node as the key node, remove the old child node in Step S134, and repeat Steps S134 - S135; if there is a collision, take the old child node in Step S134 as the new parent node and re - execute Steps S132 - S135; When the initial position coordinate is traced back, reconstruct the path planning scheme based on the key nodes and terminate the pruning optimization strategy, and extract the optimized key nodes as the sub - target points for local path planning.

[0037] Figure 3 The schematic diagram shows the key nodes in the path after pruning optimization provided by the present invention as the sub - target points for local path planning. As Figure 3 shown, during path planning, local sub - target points are sequentially selected as the current target points to construct the gravitational field. When the autonomous inspection AGV reaches the current target point, a new local sub - target point is selected as the current target point, and at the same time, the gravitational field is updated.

[0038] Through the above method, the problem that the artificial potential field algorithm is prone to fall into the U - shaped trap during path planning due to the lack of global information is avoided. The green dashed line represents the initial global path before pruning optimization, the red solid line is the final route after pruning optimization, and the local sub - target points are composed of orange nodes and the target point. If the current local sub - target point is P1, the gravitational force of the autonomous inspection AGV at point P1 acts to guide the autonomous inspection AGV to perform path planning. Through the above method, it reaches each local sub - target point in turn, and loops in this way, and finally reaches the global target point.

[0039] Step S14: Under the guidance of the sub - target points, perform local dynamic path planning based on the artificial potential field method and the dynamic gravitational field strategy.

[0040] Specifically, it includes: arranging the sub - target points in order to form a sub - target point set; in local path planning, sequentially extract the sub - target points in the sub - target point set as the current target points of the artificial potential field method; Among them, the dynamic gravitational field function is: , In the formula: is the gain coefficient; is the gravitational field gain coefficient; is the distance between the current node and the intersection point obtained by projecting the current node onto the global path, is the gravitational value of the current node.

[0041] Figure 4 is a schematic diagram of the dynamic gravitational field provided by the present invention, as shown in Figure 4 As shown, the red dashed line is the final route obtained by pruning optimization, the green solid line is the result of the route planned by the improved artificial potential field method when the autonomous inspection AGV is in a dynamic environment, and the blue dashed line is the motion trajectory of the dynamic obstacle.

[0042] When the latest node for path planning by the artificial potential field method reaches point P1, project the current latest node P1 onto the route of the global path planning. The coordinates of the corresponding dynamic gravitational field are the position where F1 is located. Among them, when the latest node of the artificial potential field method is continuously updated, the coordinates of the dynamic gravitational field are also continuously reset and updated.

[0043] Next, combined with simulation experiments, verify the solution performance of an autonomous inspection AGV dynamic path planning method. By comparing the P-RRT* algorithm, the APF-RRT algorithm with this method, solve the 50 experimental results when the dynamic obstacle acts, and record the average path length.

[0044] Figure 5 is a schematic diagram of the path planning result provided by the present invention, as shown in Figure 5As shown in the figure, two dynamic unknown obstacles are added. The obstacles move at a constant speed of 2 m / s. The yellow trajectory is the movement trajectory of the obstacles. The green dotted line is the initial route obtained by the pruning optimization strategy. The blue solid line is the route of the APF-RRT algorithm. The orange solid line is the route of the P-RRT* algorithm. The red solid line is the route of this method. From the experimental results, it can be seen that all three algorithms can find the final target point. However, in the path planning of a dynamic environment, due to the lack of the ability of dynamic obstacle avoidance, when the P-RRT* algorithm encounters dynamic unknown obstacles, the machine cannot handle them, resulting in a collision between the autonomous inspection AGV and the dynamic unknown obstacles. The APF-RRT algorithm has the ability of dynamic obstacle avoidance. When it encounters dynamic unknown obstacles, it will re-plan the path and then find the final target point. However, the local path will deviate from the result of the global route planning by a large distance, resulting in the final route length being significantly greater than the path length obtained by the pruning optimization strategy. This method also has the ability of dynamic obstacle avoidance and can find the final target point. Moreover, the local path planning result obtained by the algorithm can better fit the path result obtained by the pruning optimization strategy. The experimental data are shown in Table 1, which gives the performance parameters and operation results of three different methods. For the sake of easy expression as L avg Average path length.

[0045]

[0046] In terms of the path length, compared with the P-RRT algorithm and the APF-RRT algorithm, this method reduces by 8.31% and 3.24% on average respectively. While achieving dynamic obstacle avoidance, it effectively reduces the path length. Comprehensive analysis of the above experimental results shows that this method can be well applied to a dynamic environment. At the same time, combined with the global guidance strategy, it solves the problem that the artificial potential field method is prone to fall into the U-shaped trap and generate local optimal solutions due to the lack of global information guidance. By using the cooperative mechanism based on the dynamic gravitational field strategy, it synthesizes the results of global and local path planning, effectively reduces the path length, enables the robot to reach the target point safely and smoothly while reducing the energy consumption.

[0047] In summary, the present invention provides a method for dynamic path planning of an autonomous inspection AGV. The method includes: establishing a grid map based on environmental information, and setting the initial position coordinates and the initial target position coordinates of the autonomous inspection AGV; initializing several parameters, and implementing global path planning based on the RRT algorithm, the adaptive target biasing probability strategy, the heuristic search of the improved artificial potential field method, and the variable adaptive step function to obtain the path trajectory of the autonomous inspection AGV; optimizing the path trajectory based on the pruning optimization strategy of target backtracking, and extracting the key nodes in the pruned and optimized path trajectory as the sub-goal points for local path planning; under the guidance of the sub-goal points, performing local dynamic path planning based on the artificial potential field method and the dynamic gravitational field strategy. The present invention constructs an adaptive target biasing probability strategy. During the growth process of the random tree, by judging the environmental complexity around the parent node of the new node to be generated, that is, the area occupied by obstacles, the specific biasing probability is further determined. At the same time, an adaptive step function is constructed. During the growth process of the random tree, by judging the environmental complexity around the parent node of the new node to be generated, as well as the number of algorithm iterations and the relative distance between the random tree and the target point, the value of the adaptive step function is further determined, and higher-quality new nodes are generated during the expansion of the random tree, enabling the algorithm to plan a path with higher quality at a faster efficiency. At the same time, combined with the pruning optimization strategy of path backtracking based on the target point, redundant points in the global path are removed, further improving the path quality. The present invention constructs a method for local path planning with global key nodes as local sub-goal points in a dynamic environment, solving the problems that the target is unreachable and it is easy to fall into local optimality when the autonomous inspection AGV performs local path planning through the artificial potential field method. Through the local path algorithm optimization strategy based on the dynamic gravitational field, the problem of the coordination mechanism for integrating the global and local planning results is solved, thereby effectively reducing the path length of the autonomous inspection AGV to reach the target point.

[0048] Based on the same inventive concept, the present invention provides a device for dynamic path planning of an autonomous inspection AGV. The device includes: A map construction module for establishing a grid map based on environmental information and setting the initial position coordinates and the initial target position coordinates of the autonomous inspection AGV; An initialization module for initializing several parameters and implementing global path planning based on the RRT algorithm, the adaptive target biasing probability strategy, the heuristic search of the improved artificial potential field method, and the variable adaptive step function to obtain the path trajectory of the autonomous inspection AGV; Optimizing the path trajectory based on the pruning optimization strategy of target backtracking, and extracting the key nodes in the pruned and optimized path trajectory as the sub-goal points for local path planning; Under the guidance of sub-goal points, local dynamic path planning is carried out based on the artificial potential field method and combined with the dynamic gravitational field strategy.

[0049] Based on the same inventive concept, the present invention also provides an electronic device as shown, including: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute to implement an autonomous inspection AGV dynamic path planning method as provided above.

[0050] Based on the same inventive concept, the present invention also provides a non-transitory computer-readable storage medium. When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device can execute to implement an autonomous inspection AGV dynamic path planning method as provided above.

[0051] Since the electronic device introduced in this embodiment is the electronic device adopted for implementing the information processing method in the embodiments of the present invention, based on the information processing method introduced in the embodiments of the present invention, those skilled in the art can understand the specific implementation manners and various variations of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device implements the method in the embodiments of the present invention will not be described in detail here. As long as the electronic device adopted by those skilled in the art to implement the information processing method in the embodiments of the present invention falls within the scope of protection of the present invention.

[0052] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt 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.

[0053] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0054] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means that implement the function specified in one or more of the procedures Figure 1 or processes and / or blocks Figure 1 specified in one or more of the blocks or blocks.

[0055] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the function specified in one or more of the procedures Figure 1 or processes and / or blocks Figure 1 specified in one or more of the blocks or blocks.

[0056] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0057] Obviously, those skilled in the art can make various changes and modifications 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 equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. An autonomous inspection AGV dynamic path planning method, characterized in that, The method includes: Establishing a grid map based on environmental information, and setting the initial position coordinates and initial target position coordinates of the autonomous inspection AGV; Initializing several parameters, and implementing global path planning based on the RRT algorithm, adaptive target biasing probability strategy, heuristic search of the improved artificial potential field method, and variable adaptive step size function to obtain the path trajectory of the autonomous inspection AGV; Optimizing the path trajectory based on the pruning optimization strategy of target backtracking, and extracting the key nodes in the pruned and optimized path trajectory as sub-goal points for local path planning; Under the guidance of the sub-goal points, performing local dynamic path planning based on the artificial potential field method and dynamic gravitational field strategy.

2. The autonomous inspection AGV dynamic path planning method according to claim 1, characterized in that Establishing a grid map based on environmental information, and setting the initial position coordinates and initial target position coordinates of the autonomous inspection AGV, including: Preprocessing the environmental information, including: outlier processing, missing value processing, and feature extraction; After preprocessing, establishing a grid map according to the environmental information; In the grid map, setting the initial position coordinates and initial target position coordinates of the autonomous inspection AGV, and creating the inspection task content of the autonomous inspection AGV.

3. The dynamic path planning method for an autonomous inspection AGV according to claim 1, wherein Initializing several parameters, including: Initializing the obstacle coordinates, initial position coordinates, and initial target position coordinates of the autonomous inspection AGV; Initializing the initial step size length, target biasing probability value, safety factor of the biasing probability, gravitational gain coefficient and repulsive gain coefficient of the improved artificial potential field method of the RRT algorithm.

4. The autonomous inspection AGV dynamic path planning method according to claim 3, wherein Implementing global path planning based on the RRT algorithm, adaptive target biasing probability strategy, heuristic search of the improved artificial potential field method, and variable adaptive step size function to obtain the path trajectory of the autonomous inspection AGV, including: Constructing an adaptive target biasing probability strategy, and determining the sampling position of the random sampling points of the RRT algorithm, including: ， Among them, is an arbitrary number between 0 and 1, is the probability bias threshold, is the sampling position of the random sampling point determined according to the adaptive probability bias strategy, is the initialized target position coordinate point, is the current random sampling point obtained by the RRT algorithm; Among them, the probability bias threshold , includes: ， Among them, is the area occupied by obstacles within the circular area centered at the current coordinates of the autonomous inspection AGV and with a radius equal to the length of the current step of the autonomous inspection AGV, is the length of the current step of the autonomous inspection AGV, is the safety factor; Constructing a heuristic search based on the improved artificial potential field method, and guiding the RRT algorithm to generate random points through the improved artificial potential field method, where the gravitational field and repulsive field functions of the improved artificial potential field method include: ， ， wherein, is the gravitational field gain coefficient, is the repulsive field gain coefficient, is the distance from the current node to the target position, is the distance from the current node to the center of action of the repulsive field, is the action radius of the repulsive field, is a positive integer, is the repulsive field, is the gravitational field; The gravitational field and repulsive field functions include: ， ， ， ， Among them, is the unit vector pointing from the current node to the target point, is the unit vector pointing from the obstacle to the current node, is the gravitational value, is the repulsive force value pointing from the current node to the target point, is the value obtained by taking the partial derivative of the nth power of the distance between the current node and the target node, is the repulsive force value pointing from the obstacle to the current node, is the total repulsive force value of the current node, is the repulsive force field; Constructing a variable adaptive step size function to regulate the step size of the RRT algorithm, and the variable adaptive step size function includes: ， Among them, is the step size updated based on the adaptive step size function, is the gravitational force of the current node, is the gravitational force of the target position on the initial position, is the step size of the parent node of the current node, is the distance between the current node and the center of the repulsive force field, is the repulsive force value of the current node.

5. The autonomous inspection AGV dynamic path planning method according to claim 1, wherein Optimizing the path trajectory based on the pruning optimization strategy of target backtracking, and extracting the key nodes in the pruned and optimized path trajectory as sub-goal points for local path planning, including: Step S131, taking the target point in the result of the global path planning as the current node, and performing path backtracking based on the target point; Step S132, taking the current node as the parent node and the previous node of the current node as the sub-node, and judging whether the connection line between the parent node and the sub-node collides with an obstacle; Step S133, if the connection line between the parent node and the sub-node does not collide with an obstacle, taking the sub-node as the key node and keeping the current node still as the parent node; Step S134, taking the previous node of the sub-node as the new sub-node, and judging whether the connection line between the new sub-node and the parent node collides with an obstacle; In step S135, if no collision occurs, the new child node is taken as the key node, the old child node in step S134 is removed, and steps S134 - S135 are repeatedly executed; if a collision occurs, the old child node in step S134 is taken as the new parent node, and steps S132 - S135 are re - executed; When the initial position coordinates are traced back, a path planning scheme is reconstructed based on the key nodes, the pruning optimization strategy is terminated, and the optimized key nodes are extracted as the sub - target points for local path planning.

6. The autonomous inspection AGV dynamic path planning method according to claim 1, characterized in that, Under the guidance of the sub - target points, local dynamic path planning is carried out based on the artificial potential field method and combined with the dynamic gravitational field strategy, including: The sub - target points are arranged in order to form a sub - target point set; In local path planning, the sub - target points in the sub - target point set are sequentially extracted as the current target points of the artificial potential field method; Among them, the dynamic gravitational field function is: ， In the formula: is the gain coefficient; is the gain coefficient of the gravitational field; is the distance between the current node and the intersection point obtained by projecting the current node when it is on the global path, is the gravitational value of the current node.

7. The autonomous inspection AGV dynamic path planning method according to claim 2, wherein Missing value processing is performed on the environmental information, including: Based on the linear interpolation method, the missing values in the environmental information are processed.

8. An autonomous inspection AGV dynamic path planning device, characterized in that, The device includes: A map construction module for establishing a grid map based on the environmental information and setting the initial position coordinates and the initial target position coordinates of the autonomous inspection AGV; An initialization module for initializing several parameters and implementing global path planning based on the RRT algorithm, the adaptive target biasing probability strategy, the heuristic search of the improved artificial potential field method, and the variable adaptive step - size function to obtain the path trajectory of the autonomous inspection AGV; Based on the pruning optimization strategy of target backtracking, the path trajectory is optimized, and the key nodes in the pruned and optimized path trajectory are extracted as the sub - target points for local path planning; Under the guidance of the sub - target points, local dynamic path planning is carried out based on the artificial potential field method and combined with the dynamic gravitational field strategy.

9. An electronic device, characterized in that, Including: A processor; A memory for storing the executable instructions of the processor; Among them, the processor is configured to execute to implement an autonomous inspection AGV dynamic path planning method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device can execute to implement an autonomous inspection AGV dynamic path planning method as described in any one of claims 1 to 7.

Images