Industrial park environment robot multi-task point inspection path planning method

Through improved JPS and ant colony algorithms, and combined with TEB algorithms to process dynamic obstacles, the optimal planning and dynamic obstacle avoidance of robot patrol paths in industrial parks are achieved, and the inefficiency of path planning in traditional technology is solved. The problem has improved patrol efficiency and smooth robot movement.

CN119065375APending Publication Date: 2024-12-03HEFEI UNIV OF TECH
View PDF 0 Cites 10 Cited by

Patent Information

Application Number
CN202411190697.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

In industrial parks, traditional robot path planning technology is difficult to achieve the shortest time-consuming, optimal path and collision-free patrol routes. Especially when dynamic obstacles exist, it is difficult to avoid local optimal solutions, resulting in insufficiency of patrol.

Method used

Using the improved JPS algorithm and ant colony algorithm, the environmental data and obstacle information are obtained through sensors, the distance matrix and pheromone concentration model are constructed, and the global inspection path is optimized. When dynamic obstacles appear, the improved TEB algorithm is used for localization. Path planning, adding acceleration constraints and end smoothing constraints to improve motion smoothness.

Benefits of technology

It realizes the optimal planning of robot patrol paths in industrial parks, shortens the path planning time, reduces the path length, improves the smoothness of robot motion and motor life, avoids sudden collisions of dynamic obstacles, and ensures patrol inspection. The task was completed smoothly.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119065375A_ABST
    Figure CN119065375A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-task-point inspection path planning method for an industrial park environment robot, and the method comprises the steps: obtaining the cost value of any two target points based on an improved JPS, obtaining a distance matrix, and obtaining an optimal inspection sequence of inspection tasks through an ant colony algorithm. The improved JPS algorithm is used for global path planning, the global optimal inspection path is obtained, the path planning time is shortened, the path length is reduced, and the obtained path is more suitable for industrial environment inspection. Obstacle avoidance of dynamic obstacles is realized in a global path elastic stretching mode, and turning points of the path are reduced; the added jerk constraint prevents the impact oscillation of the robot, increases the smoothness of movement and prolongs the service life of the motor. According to the invention, the problem of multi-task point inspection of the mobile robot in the industrial park can be solved, a smooth path with short time consumption can be obtained, dynamic obstacles can be avoided in real time, the problem of falling into a local optimal solution can be effectively avoided, and the target point can be quickly reached to complete the inspection task.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of multi-target point path planning for mobile robots, and in particular to a multi-task point inspection path planning method for an industrial park environment robot. Background Art

[0002] In recent years, robot path planning technology has been widely used in various fields, playing a vital role in daily life, industrial production, protection and rescue. Especially in industrial scenarios, in order to ensure the normal operation of equipment and the safety of workers, it is necessary to inspect and maintain the on-site environment. In traditional technology, industrial parks use manual inspections, which often threaten the safety of workers, damage their health, and increase the labor cost of inspection tasks. Inspection robots, as a new technology, have replaced manual inspections and have received widespread attention. Path planning, as part of the robot's mission, is the premise and guarantee for the robot to complete its inspection tasks.

[0003] Inspection robots in industrial parks need to complete a large number of inspection tasks, which requires the robots to start from the starting point and arrive at different task points in sequence until they traverse all task points and return to the starting point after completing the work tasks. In this process, in addition to completing the operations at the task points, the entire traversal path must be the shortest and must not collide with any obstacles, so that the entire process takes a short time. Therefore, for mobile robot inspections in industrial parks, planning the shortest, optimal and collision-free routes so that they can complete various tasks in an orderly manner plays a significant role in reducing the production and maintenance costs of enterprises. Summary of the invention

[0004] The purpose of the present invention is to make up for the defects of the existing technology and provide a multi-task point inspection path planning method for robots in an industrial park environment to solve the problem of multi-task point inspection of robots in industrial scenarios, to achieve the shortest time, the best path and the collision-free route, and to complete the inspection tasks in an orderly manner.

[0005] The present invention is achieved through the following technical solutions:

[0006] A method for multi-task inspection path planning of an industrial park environment robot specifically comprises the following steps:

[0007] Step 1: Obtain environmental data through the sensor equipment carried by the robot, establish an industrial park scene map, and obtain obstacle information and inspection target point location coordinates;

[0008] Step 2: Use the JPS algorithm to calculate the cost between any two inspection target points and construct a distance matrix;

[0009] Step 3: According to the inspection target points and the inspection cost values ​​of each inspection point, the optimal global inspection target point sequence is obtained based on the ant colony algorithm;

[0010] Step 4: Based on the optimal global inspection target point sequence, use the JPS algorithm to plan the path of adjacent points to obtain the global optimal inspection path;

[0011] Step 5: Detect obstacles in real time and update the environment map; when no dynamic obstacles are detected, perform inspections according to the global optimal inspection path; when dynamic obstacles are detected, use the TEB algorithm for local path planning and perform inspections according to the path planned by the local path.

[0012] In step 1, the surrounding environment information is obtained through the multi-line laser radar carried by the mobile robot, raster map modeling is performed, obstacle information and the coordinates of all target points are obtained, and obstacles are processed to finally determine the freely traversable areas and inaccessible areas in the map.

[0013] The JPS algorithm described in step 2 is used to calculate the cost value between any two inspection target points and construct a distance matrix, as follows: In view of the low efficiency of the heuristic function of the traditional JPS algorithm, a new heuristic function is proposed based on diagonal distance and goal orientation:

[0014] h(n)=iL-jP(α)

[0015] Among them, i and j are the weights of distance information and goal orientation respectively, L is the diagonal distance estimate from the current node to the target node, P(α) is the goal orientation function, and the calculation formula of L is:

[0016]

[0017] In the formula, (x n ,y n ) and (x g ,y g ) represent the coordinates of the current node n and the target node g respectively;

[0018] The goal-oriented function P(α) adds the target direction information to enhance the inspiration. n and P g The direction vector between Then define the goal-oriented function in direction α:

[0019]

[0020] In the formula, is the direction vector in the direction α;

[0021] The JPS algorithm is implemented as follows: 1) add the starting point s to the search node S(n); 2) g(n) is the actual distance from the starting point to the current node n, h(n) is the heuristic function, and based on the evaluation function f(n)=g(n)+h(n) using the heuristic function, calculate the cost of all search nodes S(n), record the minimum cost node as n, set n as the current node, and determine whether n is the end point. If yes, start from the end point and trace back to the parent node of each node in the temporary path T(n) in turn until the starting point, and record the path composed of the backtracked nodes. If the path is P, execute 7), otherwise execute 3); 3) Is the current node s? If so, execute 4), otherwise execute 5); 4) Search for jump points in the horizontal, vertical and diagonal directions along all directions, and execute 6); 5) According to the direction of the parent node, search for jump points in the horizontal, vertical and diagonal directions, and execute 6); 6) For the parent node of the jump point found, add the parent node to the temporary path T(n), and execute 2); 7) Output the final path P; Any two target points can obtain the optimal path in this way, and the length of the path is used as the cost value, and finally construct the distance matrix.

[0022] According to the inspection target points and the inspection cost values ​​of each inspection point, the optimal global inspection target point sequence is obtained based on the ant colony algorithm as described in step 3, specifically including: 1) obtaining the distance matrix between the inspection points; 2) initializing the ant colony scale, pheromone importance factor, heuristic function importance factor, pheromone volatility factor, pheromone constant, maximum number of iterations parameters, and the initial placement of ants; 3) each ant selects the next inspection point according to the transfer probability and updates the pheromone on the path; 4) after the ant walks, the pheromone concentration is updated according to the path length; 5) the global optimal path is recorded, and the strategy is adjusted according to the information concentration; 6) the search is terminated when the preset number of iterations is reached or specific conditions are met, and the optimal path is returned.

[0023] The transition probability formula is as follows:

[0024]

[0025] Among them, α and β are regulatory factors, τ ij (t) is the pheromone concentration, η ij is the distance from the current point to the next target point, allowed k represents the path that ant k has not walked yet; the pheromone formula is as follows:

[0026]

[0027] Among them, ρ is the pheromone volatility factor, m is the total number of ants, The pheromones left by all ants passing through i to j target point.

[0028] Based on the optimal global inspection target point sequence described in step 4, the JPS algorithm is used to plan the path of adjacent points to obtain the global optimal inspection path, which is as follows: the JPS algorithm finds the shortest path between two adjacent target points according to the process described in step 2; after traversing all target points, a global optimal inspection path from the starting point to the end point is found.

[0029] In step 5, when a dynamic obstacle is detected, the TEB algorithm is used to obtain the local path. Based on the original TEB algorithm, the jerk constraint is added to limit the rate of change of the jerk to a smaller range, improve the smoothness of the robot's movement and increase the life of the motor; in order to prevent the mobile robot from suddenly jumping from the maximum speed to a standstill when arriving at the inspection point, causing a large impact on itself, and sliding after reaching the target point, an end smoothing constraint is added to adjust the robot's speed when the robot reaches the target area, so that it can slow down and reach the target point smoothly. , as follows:

[0030] Add the acceleration constraint to the hypergraph and construct a new hypergraph. The edge j0 is connected to the three time interval vertices ΔT0, ΔT1, and ΔT2 and the four posture vertices s0, s1, s2, and s3. The number of time intervals and postures connected by j1 remains unchanged and is postponed by one sequence number respectively.

[0031] s0, s1, s2, s3, s4 are the five consecutive posture states of the robot, v0, v1, v2, v3 are the velocity states (including linear velocity and angular velocity) connecting the five postures, a0, a1, a2 are the acceleration states (including linear acceleration and angular acceleration), a 1lin 、a 0lin They represent linear acceleration, a 1rot 、a 0rot They represent angular acceleration respectively. The jerk constraints satisfied by j0 and j1 are as follows:

[0032] The linear acceleration is expressed as

[0033]

[0034] The angular acceleration is expressed as

[0035]

[0036] Adding a terminal smoothing constraint adjusts the robot speed when it reaches the target area, allowing it to slow down and reach the target point smoothly, while reducing the impact of the robot and achieving more accurate speed planning.

[0037]

[0038] Where: v maxis the preset maximum linear velocity of the robot, d threshold is the pre-set threshold of the diagonal distance between the robot and the target point, d is the diagonal distance between the robot's current position and the target point, v dmax is the maximum linear speed of the robot during movement;

[0039] Construct the objective function f(B) of the TEB algorithm with multiple constraints, use the g2o open source graph optimization library to solve the objective function, and finally obtain a smooth local path that satisfies the constraints;

[0040] Finally, the robot multi-task point inspection path planning method performs inspections according to the planned optimal path and uses the TEB algorithm to avoid obstacles.

[0041] The advantages of the present invention are: the present invention aims at the multi-task point inspection in the industrial park, obtains the cost value of any two target points based on the improved JPS algorithm, obtains the distance matrix, and obtains the inspection order of the inspection task through the ant colony algorithm. The present invention uses the improved JPS algorithm to perform global path planning based on the obtained optimal inspection order, obtains the global optimal inspection path, and shortens the path planning time and reduces the path length compared with the A* algorithm. The obtained path is more suitable for industrial environment inspection. The present invention uses the improved algorithm TEB, combined with the global path point posture of the robot, to achieve dynamic obstacle avoidance in the form of global path elastic stretching, and reduces the turning point of the path; the added acceleration constraint prevents the impact oscillation of the robot, increases the smoothness of the movement and improves the life of the motor; adds the terminal smooth constraint, so that when the robot reaches the target area, the robot speed is adjusted to slow down and reach the target point smoothly. The present invention can solve the problem of multi-task point inspection of mobile robots in industrial parks, can obtain a short and smooth path, can avoid dynamic obstacles in real time, effectively avoid the problem of falling into the local optimal solution, and achieve rapid arrival at the target point to complete the inspection task. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of the multi-task point inspection path planning method of the industrial park environment robot of the present invention.

[0043] Figure 2 This is the flow chart of the improved JPS algorithm of the present invention

[0044] Figure 3 It is a g2o hypergraph model diagram constructed by the present invention. DETAILED DESCRIPTION

[0045] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0046] like Figure 1As shown, this embodiment discloses a multi-task inspection path planning method for an industrial park environment robot, comprising the following steps:

[0047] Step 1: Obtain environmental data through the sensor equipment carried by the robot, establish an industrial park scene map, and obtain obstacle information and inspection target point location coordinates.

[0048] The multi-line laser radar carried by the mobile robot obtains information about the surrounding environment, performs grid map modeling, determines the coordinates of obstacles and all target points, and performs obstacle processing on the obstacles, ultimately determining the freely traversable areas and inaccessible areas in the map.

[0049] Step 2: Use the improved JPS algorithm to calculate the cost between any two inspection target points and construct a distance matrix.

[0050] Aiming at the low efficiency of the heuristic function of the traditional JPS algorithm, a new heuristic function is proposed based on diagonal distance and goal orientation as shown in formula (1):

[0051] h(n)=iL-jP(α) (1)

[0052] Where i and j are the weights of distance information and target orientation, respectively. L is the estimated diagonal distance from the current node to the target node. The calculation formula of L is shown in formula (2):

[0053]

[0054] In the formula, (x n ,y n ) and (x g ,y g ) represent the coordinates of the current node n and the target node g respectively;

[0055] The goal-oriented function P(α) adds the target direction information to enhance the inspiration, using P n and P g The direction vector between Then the goal-oriented function in direction α is defined:

[0056]

[0057] In the formula, is the direction vector in the direction α;

[0058] like Figure 2As shown in the figure, the improved JPS algorithm is implemented as follows: 1) add the starting point s to the search node S(n); 2) g(n) is the actual distance from the starting point to the current node n, h(n) is the heuristic function, based on the evaluation function f(n)=g(n)+h(n) using the new heuristic function, calculate the cost of all search nodes S(n), let the minimum cost node be n, set n as the current node, and determine whether n is the end point. If so, start from the end point and trace back in the temporary path T(n) to output the parent node of each node in turn until the starting point, and record the backtracking node composition If the path is P, execute 7), otherwise execute 3); 3) Is the current node s? If yes, execute 4), otherwise execute 5); 4) Search for jump points in the horizontal, vertical, and diagonal directions along all directions, and execute 6); 5) Search for jump points in the horizontal, vertical, and diagonal directions according to the direction of the parent node, and execute 6); 6) For the parent node of the jump point found, add the parent node to the temporary path T(n), and execute 2); 7) Output the final path P; Any two target points can obtain the optimal path in this way, and the length of the path is used as the cost value, and finally construct the distance matrix D. As shown in formula (4)

[0059]

[0060] Step 3: According to the inspection target point information and the inspection cost of each inspection point, the optimal global inspection target point sequence is obtained based on the ant colony algorithm.

[0061] According to the cost value between the inspection points, based on the ant colony algorithm, the optimal global inspection target point sequence is obtained, which specifically includes: 1) obtaining the distance matrix between the inspection points; 2) initializing the parameters such as the ant colony scale, pheromone important factor, heuristic function important factor, pheromone volatility factor, pheromone constant, maximum number of iterations, and the initial placement of ants; 3) each ant selects the next city according to the transfer probability and updates the pheromone on the path; 4) after the ant walks, the pheromone concentration is updated according to the path length; 5) recording the global optimal path and adjusting the strategy according to the information concentration; 6) ending the search when the preset number of iterations is reached or specific conditions are met, and returning to the optimal path;

[0062] The transition probability formula is shown in formula (5):

[0063]

[0064] Among them, α and β are regulatory factors, τ ij (t) is the pheromone concentration, η ij is the distance from the current point to the next target point, allowed k represents the path that ant k has not walked yet; the pheromone formula is shown in formula (6):

[0065]

[0066] Among them, ρ is the pheromone volatility factor, m is the total number of ants, The pheromones left by all ants passing through i to the target point.

[0067] Step 4: Based on the optimal global inspection target point sequence, use the improved JPS algorithm to plan the path of adjacent points to obtain the global optimal inspection path.

[0068] According to the global optimal inspection target point sequence obtained by the ant colony algorithm, the improved JPS algorithm is used to plan the path of adjacent points to obtain the global optimal inspection path; the improved JPS algorithm can find the shortest path between two adjacent target points according to the process described in step 2; after traversing all target points, a global optimal inspection path from the starting point to the end point can be found.

[0069] Step 5: Detect obstacles in real time and update the environment map. When no dynamic obstacles are detected, perform inspections according to the global optimal inspection path. When dynamic obstacles are detected, use the improved TEB algorithm for local path planning. Perform inspections according to the path planned by the local path.

[0070] When a dynamic obstacle is detected, the improved TEB algorithm is used to obtain the local path. Based on the original TEB algorithm, the jerk constraint is added to limit the rate of change of jerk to a smaller range to prevent the robot from being impacted by excessive acceleration, thereby obtaining more accurate speed planning, improving the smoothness of the robot's movement, and increasing the life of the motor. Figure 3 As shown in the figure, the jerk constraint is added to the hypergraph to construct a new hypergraph. The edge j0 is connected to the three time interval vertices ΔT0, ΔT1, and ΔT2 and the four posture vertices s0, s1, s2, and s3. The time intervals and postures connected by j1 remain unchanged and are postponed by one sequence number respectively.

[0071] s0, s1, s2, s3, s4 are the five consecutive posture states of the robot, v0, v1, v2, v3 are the velocity states (including linear velocity and angular velocity) connecting the five postures, a0, a1, a2 are the acceleration states (including linear acceleration and angular acceleration), a 1lin 、a 0lin They represent linear acceleration, a 1rot 、a 0rot They represent angular acceleration respectively. The jerk constraints satisfied by j0 and j1 are as follows:

[0072] The linear acceleration constraint can be expressed as

[0073]

[0074] The angular acceleration constraint can be expressed as

[0075]

[0076] In the process of program implementation, it is necessary to redefine the jerk edge as a constraint condition and add the jerk constraint to the g2o library. It is applied in the starting state, intermediate state and terminal state of the local planning respectively.

[0077] Adding a terminal smoothing constraint adjusts the robot speed when it reaches the target area, allowing it to slow down and reach the target point smoothly, while reducing the impact of the robot and achieving more accurate speed planning.

[0078]

[0079] Where: v max is the preset maximum linear velocity of the robot, d threshold is the pre-set threshold of the diagonal distance between the robot and the target point, d is the diagonal distance between the robot's current position and the target point, v dmax It is the maximum linear speed of the robot during movement.

[0080] The traditional TEB algorithm includes obstacle constraints, path point targets, non-holonomic kinematic constraints, speed and acceleration constraints, and time-optimal objective function constraints. On this basis, jerk constraints and terminal smoothness constraints are added to construct the objective function f(B) of the TEB algorithm with multiple constraints. The g2o open source graph optimization library is used to solve the objective function and finally obtain a smooth local path that satisfies the constraints.

[0081] Finally, the robot multi-task point inspection path planning method performs inspections according to the planned optimal path and uses the improved TEB algorithm to avoid obstacles.

[0082] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. The embodiments described in the present invention are merely descriptions of the preferred embodiments of the present invention, and do not limit the concept and scope of the present invention. The various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction, and such combinations should also be regarded as the contents disclosed by the present invention as long as they do not violate the concept of the present invention. In order to avoid unnecessary repetition, the present invention will not further explain various possible combinations.

[0083] The present invention is not limited to the specific details of the above-mentioned embodiments. Within the scope of the technical concept of the present invention and without departing from the design concept of the present invention, various modifications and improvements made to the technical solution of the present invention by technical personnel in this field should fall within the protection scope of the present invention. The technical contents for which protection is sought in the present invention have been fully recorded in the claims.

Claims

1. A multi-task inspection path planning method for an industrial park environment robot, characterized in that: The specific steps include: Step 1: Obtain environmental data through the sensor equipment carried by the robot, establish an industrial park scene map, and obtain obstacle information and inspection target point location coordinates; Step 2: Use the JPS algorithm to calculate the cost between any two inspection target points and construct a distance matrix; Step 3: According to the inspection target points and the inspection cost values ​​of each inspection point, the optimal global inspection target point sequence is obtained based on the ant colony algorithm; Step 4: Based on the optimal global inspection target point sequence, use the JPS algorithm to plan the path of adjacent points to obtain the global optimal inspection path; Step 5: Detect obstacles in real time and update the environment map; when no dynamic obstacles are detected, perform inspections according to the global optimal inspection path; when dynamic obstacles are detected, use the TEB algorithm for local path planning and perform inspections according to the path planned by the local path.

2. The method for multi-task inspection path planning of an industrial park environment robot according to claim 1 is characterized in that: In step 1, the surrounding environment information is obtained through the multi-line laser radar carried by the mobile robot, raster map modeling is performed, obstacle information and the coordinates of all target points are obtained, and obstacles are processed to finally determine the freely traversable areas and inaccessible areas in the map.

3. The method for multi-task inspection path planning of an industrial park environment robot according to claim 1 is characterized in that: The JPS algorithm described in step 2 is used to calculate the cost value between any two inspection target points and construct a distance matrix, as follows: Based on the diagonal distance and target orientation, a heuristic function is proposed: h(n)=iL-jP(α) Among them, i and j are the weights of distance information and goal orientation respectively, L is the diagonal distance estimate from the current node to the target node, P(α) is the goal orientation function, and the calculation formula of L is: In the formula, (x n ,y n ) and (x g ,y g ) represent the coordinates of the current node n and the target node g respectively; The goal-oriented function P(α) adds the target direction information to enhance the inspiration. n and P g The direction vector between Then define the goal-oriented function in direction α: In the formula, is the direction vector in the direction α; The JPS algorithm is implemented as follows: 1) add the starting point s to the search node S(n); 2) g(n) is the actual distance from the starting point to the current node n, h(n) is the heuristic function, and based on the evaluation function f(n)=g(n)+h(n) using the heuristic function, calculate the cost of all search nodes S(n), record the minimum cost node as n, set n as the current node, and determine whether n is the end point. If yes, start from the end point and trace back to the parent node of each node in the temporary path T(n) in turn until the starting point, and record the path composed of the backtracked nodes. If the path is P, execute 7), otherwise execute 3); 3) Is the current node s? If so, execute 4), otherwise execute 5); 4) Search for jump points in the horizontal, vertical and diagonal directions along all directions, and execute 6); 5) According to the direction of the parent node, search for jump points in the horizontal, vertical and diagonal directions, and execute 6); 6) For the parent node of the jump point found, add the parent node to the temporary path T(n), and execute 2); 7) Output the final path P; Any two target points can obtain the optimal path in this way, and the length of the path is used as the cost value, and finally construct the distance matrix.

4. The method for multi-task inspection path planning of an industrial park environment robot according to claim 1 is characterized in that: According to the inspection target points and the inspection cost values ​​of each inspection point, the optimal global inspection target point sequence is obtained based on the ant colony algorithm as described in step 3, specifically including: 1) obtaining the distance matrix between the inspection points; 2) initializing the ant colony scale, pheromone importance factor, heuristic function importance factor, pheromone volatility factor, pheromone constant, maximum number of iterations parameters, and the initial placement of ants; 3) each ant selects the next inspection point according to the transfer probability and updates the pheromone on the path; 4) after the ant walks, the pheromone concentration is updated according to the path length; 5) the global optimal path is recorded, and the strategy is adjusted according to the information concentration; 6) the search is terminated when the preset number of iterations is reached or specific conditions are met, and the optimal path is returned.

5. The method for multi-task inspection path planning of an industrial park environment robot according to claim 4 is characterized in that: The transition probability formula is as follows: Among them, α and β are regulatory factors, τ ij (t) is the pheromone concentration, η ij is the distance from the current point to the next target point, allowed k represents the path that ant k has not walked yet; the pheromone formula is as follows: Among them, ρ is the pheromone volatility factor, m is the total number of ants, The pheromones left by all ants passing through i to j target point.

6. The method for multi-task inspection path planning of an industrial park environment robot according to claim 1, characterized in that: Based on the optimal global inspection target point sequence described in step 4, the JPS algorithm is used to plan the path of adjacent points to obtain the global optimal inspection path, which is as follows: the JPS algorithm finds the shortest path between two adjacent target points according to the process described in step 2; after traversing all target points, a global optimal inspection path from the starting point to the end point is found.

7. The method for multi-task inspection path planning of an industrial park environment robot according to claim 1, characterized in that: When a dynamic obstacle is detected as described in step 5, the TEB algorithm is used for local path planning, and inspection is performed according to the path planned by the local path, as follows: Add the acceleration constraint to the hypergraph and construct a new hypergraph. The edge j0 is connected to the three time interval vertices ΔT0, ΔT1, and ΔT2 and the four posture vertices s0, s1, s2, and s3. The number of time intervals and postures connected by j1 remains unchanged and is postponed by one sequence number respectively. s0, s1, s2, s3, s4 are the five consecutive posture states of the robot, v0, v1, v2, v3 are the velocity states connecting the five postures, including linear velocity and angular velocity, a0, a1, a2 are the acceleration states, including linear acceleration and angular acceleration, a 1lin 、a 0lin They represent linear acceleration, a 1rot 、a 0rot They represent angular acceleration respectively. The jerk constraints satisfied by j0 and j1 are as follows: The linear acceleration constraint is expressed as The angular acceleration constraint is expressed as Adding a terminal smoothing constraint adjusts the robot speed when it reaches the target area, allowing it to slow down and reach the target point smoothly, while reducing the impact of the robot and achieving more accurate speed planning. Where: v max is the preset maximum linear velocity of the robot, d threshold is the pre-set threshold of the diagonal distance between the robot and the target point, d is the diagonal distance between the robot's current position and the target point, v dmax is the maximum linear speed of the robot during movement; Construct the objective function f(B) of the TEB algorithm with multiple constraints, use the g2o open source graph optimization library to solve the objective function, and finally obtain a smooth local path that satisfies the constraints; Finally, the robot multi-task point inspection path planning method performs inspections according to the planned optimal path and uses the TEB algorithm to avoid obstacles.

Citation Information

Cited By

  • Inspection robot path planning method based on transformer substation environment

    CN119374602A

  • Sampling task multi-region path planning method and system based on time limit constraint segmentation

    CN119809073A

  • Robot cluster scheduling method, device and equipment

    CN119902446A

  • Well lid intelligent inspection path dynamic planning method, vehicle-mounted well lid intelligent inspection device and inspection vehicle

    CN120628132A

  • Intelligent inspection path dynamic planning method for well covers, vehicle-mounted intelligent inspection device for well covers, and inspection vehicle

    CN120628132B