Intelligent inspection task allocation and path optimization method for multi-heterogeneous robots in chemical industry park

Abstract

The invention discloses a chemical industry park-oriented multi-heterogeneous robot inspection task allocation and path optimization intelligent method, and belongs to the field of multi-heterogeneous robot chemical inspection. Comprising the following steps: step 1, extracting environment information such as the position of a to-be-inspected point and the requirement of the position of the to-be-inspected point on the passing ability of a robot in a chemical industry park, so as to subsequently process data; 2, establishing a mapping relation between a coding mode and a solution space by setting a two-layer coding mode; step 3, taking the minimum total path cost of the robot for completing the distributed tasks as an optimization target, and using an improved mixed artemisinin optimization algorithm to optimize the distribution of the inspection tasks and the inspection route of the robot; and 4, information such as distribution of robot inspection tasks and path optimization results is visualized. According to the method, effective technical support is provided for reasonable distribution of the inspection tasks and planning of the optimal inspection route for the multi-heterogeneous robots in the chemical industry park, the multi-robot inspection efficiency is improved, the safety of the chemical industry park is maintained, and the robot inspection cost is reduced.

Description

Technical Field

[0001] The present invention belongs to the field of multi-heterogeneous robot chemical safety inspection, and specifically relates to an intelligent method for multi-heterogeneous robot inspection task allocation and path optimization for chemical parks. Background Art

[0002] Petrochemical enterprises have large-scale equipment and complex processes. For complex spatial conditions, traditional manual inspections are prone to problems such as false inspections and missed inspections. More importantly, they pose safety risks to inspectors.

[0003] With the development of technologies such as artificial intelligence and robots, it has become possible for inspection robots to replace traditional manual inspections of dangerous factors such as "leakage, bubbling, dripping, and leakage" in chemical parks, abnormal equipment temperature, and abnormal sounds. There are various terrains in chemical parks, such as flat ground, stairs, and pipelines, and the inspection tasks are distributed in different terrain areas in the park, making it difficult for single-type robots (including multiple single-type robots) to complete the inspection tasks. In order to quickly identify dangerous factors in chemical parks and improve inspection efficiency, it is very necessary to use multiple heterogeneous robots to collaboratively complete the detection of dangerous factors in chemical parks.

[0004] However, there are relatively few studies on robot inspection route optimization for chemical parks, and they are mainly based on single robots, without considering the complex terrain environment of chemical parks and the applicability of the terrain of different inspection locations to the robot's ability to pass. Therefore, in order to address these problems, it is very necessary to study a multi-heterogeneous robot task allocation scheme that considers the robot's ability to pass, and a robot path optimization method. Summary of the invention

[0005] In view of the deficiencies of the above-mentioned prior art, the present invention proposes an intelligent method for multi-heterogeneous robot inspection task allocation and path optimization for chemical parks. A two-layer encoding method is set to establish a mapping relationship between the encoding method and the solution space, and the complex terrain environment of the chemical park and the applicability of the terrain of different inspection locations to the robot's passing ability are taken into consideration. The total distance cost of the inspection robot is minimized as the optimization goal, and an improved mixed artemisinin optimization algorithm is used to allocate inspection tasks to the robots, and an optimal inspection route is planned for each robot, thereby improving the efficiency of multiple robots in performing chemical park inspection tasks. Effective technical support is provided for the reasonable allocation of inspection tasks and planning of optimal inspection routes for multiple heterogeneous robots in chemical parks, which effectively improves the efficiency of inspection and maintains the safety of chemical parks. It has important practical value in the field of multi-robot inspection for chemical parks.

[0006] To achieve the above object, the present invention is implemented by the following technical solutions:

[0007] The intelligent method for multi-heterogeneous robot inspection task allocation and path optimization for chemical parks has the following specific steps:

[0008] Step 1: Extract environmental information of chemical parks;

[0009] Step 2: Establish a mapping relationship between the encoding method and the solution space by setting a two-layer encoding method;

[0010] Step 3: After the two-layer encoding method establishes the mapping relationship between the encoding method and the solution space, the improved hybrid artemisinin optimization algorithm is used to assign appropriate inspection tasks to the robot and optimize the inspection route;

[0011] Step 4: Use MATLAB simulation software to visualize information such as the allocation of robot inspection tasks in the chemical park and path optimization results.

[0012] Further, the step 1 is specifically as follows: the chemical park environmental information includes the inspection point location, the inspection point location requirements for the robot inspection capability, the robot warehouse location, the chemical park route and obstacles, etc., and the chemical park environmental information is stored in an Excel table; then the table data is read using MATLAB simulation software, and A * The algorithm calculates the route trajectory and journey cost between each inspection point, providing data for subsequent robot inspection task allocation and route optimization.

[0013] Further, the step 2 is specifically as follows:

[0014] After using MATLAB simulation software to read the Excel table storing the environmental information of the chemical park, a two-layer coding method is set to establish a spatial mapping relationship between the coding method and the inspection tasks assigned to the robot and the order in which the inspection tasks are executed. This problem includes two sub-problems: assigning inspection points to the robot and sorting the order in which the robot executes the inspection points. A two-layer coding based on random keys is used for each individual, in which each layer has the same length. The coding form can be described as:

[0015]

[0016] Among them, the first layer is used to indicate the priority order of the inspection points to be executed by the robot, and the second layer is used to indicate the robot number assigned to the inspection points; each element in X can take any value in [-ε,ε]; the value of ε can take any appropriate integer, and l represents the number of inspection points in the chemical park;

[0017] Encode the formula (1) to map the space of the order of assigning patrol tasks to robots and the order of robots executing patrol tasks. According to formula (2), convert the element values of the individual position vector into the serial numbers in the set of robots that can be selected for the positions to be patrolled. Thus, the robot number for executing the task at the position to be patrolled can be obtained;

[0018]

[0019] Among them, floor represents the floor operation, i represents the number of the position to be patrolled, s i represents the number of robots that can be selected for the position with the number i of the position to be patrolled, z i ∈[1, s i represents the serial number of the robot assigned to the position to be patrolled i in the set of robots that can be selected for this patrol point, that is, the z i th element in the set of robots that can be assigned to the position to be patrolled i is the robot number for executing the task at this position to be patrolled;

[0020] Convert the encoding form to the order of robots executing patrol tasks. For each position to be patrolled assigned by the above formula (2), sort the first-layer elements of the positions to be patrolled assigned to the robots with the same number in ascending order, and use the sorting size to represent the order of robots executing the tasks at each position to be patrolled.

[0021] Furthermore, the specific content of step three is as follows:

[0022] In the MATLAB simulation software, use the improved hybrid artemisinin optimization algorithm to solve the optimal route for each robot to start from the warehouse position, execute all assigned patrol tasks, and finally return to the robot warehouse;

[0023] 1) Introduce the Sigmoid function in the comprehensive clearance stage of the artemisinin optimization algorithm to improve the drug concentration factor C, thereby enhancing the global search ability of the artemisinin optimization algorithm. Simulate the initial stage of artemisinin drug treatment for malaria, and introduce the probability factor K. When rand < K, the movement position of the artemisinin drug particle is described as:

[0024]

[0025] K = 1 - t 1 / 6 / t max 1 / 6 (4)

[0026]

[0027] In the formula: and are the artemisinin positions before and after update respectively, is the current optimal position of artemisinin, t, t maxare the current iteration number and the maximum iteration number, respectively, and c represents the drug concentration factor;

[0028] 2) In order to further balance the global search ability and local search ability of the artemisinin optimization algorithm, in the local clearance stage, the walking optimization algorithm is introduced to update the movement position of artemisinin drug particles and improve the local search ability of the artemisinin algorithm. The walking optimization algorithm is described as follows:

[0029] s=tan(theta) (6)

[0030] vel=6e -3.5(s+0.05) (7)

[0031]

[0032] Where theta represents a random number in the set [0,1,2…50], which is used to simulate the trajectory or terrain, SF represents any number between 1 and 2, s represents the inclination angle of the trajectory or terrain, and vel represents the speed of the hiker;

[0033] 3) Late consolidation stage: In the late consolidation stage, patients are still likely to have relapses of malaria; therefore, the movement position of artemisinin drug particles in the late consolidation stage is updated as follows:

[0034]

[0035] 4) Repeatedly improve the mixed artemisinin optimization algorithm until the maximum number of iterations is completed, and select the best individual as the optimal task allocation and inspection route plan for the robot.

[0036] Furthermore, the step four is specifically as follows: using MATLAB simulation software, the inspection points assigned to each robot obtained by solving the steps one to three, as well as the optimal path for each robot to start from the warehouse, perform all assigned inspection tasks and finally return to the robot warehouse are visualized, thereby facilitating use by actual chemical enterprise staff.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] The present invention proposes an intelligent method for allocating inspection tasks and optimizing paths for multi-heterogeneous robots in chemical parks. The method takes into account the complex terrain environment of chemical parks and the applicability of terrains at different locations to be inspected to the robot's ability to pass through. By setting a two-layer encoding method, a mapping relationship between the encoding method and the solution space is established. The optimization goal is to minimize the total cost of the inspection robot. The improved mixed artemisinin optimization algorithm is used to optimize the task allocation scheme and the robot inspection route, thereby improving the efficiency of multi-robot inspection tasks in chemical parks. The method provides effective technical support for the rational allocation of inspection tasks and planning of optimal inspection routes for multi-heterogeneous robots in chemical parks, effectively improves the efficiency of inspections, and maintains the safety of chemical parks. It has important practical value in the field of multi-heterogeneous robot inspections for chemical parks. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a flow chart of the intelligent method for multi-heterogeneous robot inspection task allocation and path optimization for chemical parks of the present invention;

[0040] Figure 2 This is the environmental map of the chemical park inspection mission;

[0041] Figure 3 The convergence curve diagram of the optimal inspection route is solved for the two-layer coding method and the hybrid artemisinin optimization algorithm;

[0042] Figure 4 Route planning map for multiple heterogeneous robots to complete inspection tasks in chemical parks. DETAILED DESCRIPTION

[0043] Below is the attached figure Figures 1 to 4 The present invention is further explained with reference to the accompanying drawings and embodiments.

[0044] The following will be combined with the benchmark data set and the accompanying drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0045] Intelligent method for multi-heterogeneous robot inspection task allocation and path optimization for chemical parks. The overall process is as follows Figure 1 The actual specific steps are as follows:

[0046] Step 1: Extract environmental information of chemical parks;

[0047] The information such as the location of inspection points in the chemical park, the requirements of the inspection point location on the robot inspection capability, the location of the robot warehouse, the route of the chemical park and obstacles are stored in an Excel table; then the table data is read using MATLAB simulation software, and A *The algorithm calculates the route trajectory and journey cost between each inspection point, providing data for subsequent robot inspection task allocation and route optimization;

[0048] Specifically, the location of inspection points in the chemical park, the requirements of the inspection point location on the robot inspection capability, the location of the robot warehouse, the route of the chemical park, and obstacles are recorded; A * The algorithm calculates the route trajectory and route cost between each inspection point, providing data for subsequent robot inspection task allocation and route optimization. The MATLAB simulation software is used to draw the chemical park inspection environment information such as Figure 2 As shown in the figure, the black part represents the obstacles in the chemical park, the five-pointed star in the lower left corner represents the robot warehouse, which is the starting point and end point of the robot. The inspection point represented by the circle can only be reached and completed by a quadruped robot with stronger passing ability due to the terrain environment. The inspection point represented by the triangle has a flat terrain, and all robots can reach and complete the inspection task of the inspection point.

[0049] Step 2: Establish a mapping relationship between the encoding method and the solution space by setting a two-layer encoding method;

[0050] After using MATLAB simulation software to read the Excel table storing the environmental information of the chemical park, a two-layer coding method is set to establish a spatial mapping relationship between the coding method and the inspection tasks assigned to the robot and the order in which the inspection tasks are executed. This problem includes two sub-problems: assigning inspection points to the robot and sorting the order in which the robot executes the inspection points. A two-layer coding based on random keys is used for each individual, in which each layer has the same length. The coding form can be described as:

[0051]

[0052] Among them, the first layer is used to indicate the priority order of the inspection points to be executed by the robot, and the second layer is used to indicate the robot number assigned to the inspection points; each element in X can take any value in [-ε,ε]; the value of ε can take any appropriate integer, and l represents the number of inspection points in the chemical park;

[0053] The encoding of formula (1) is mapped to the space of assigning inspection tasks to robots and the order of robots executing inspection tasks. According to formula (2), the element value of the individual position vector is converted into the serial number of the optional robot set at the position to be inspected, thereby obtaining the robot number that executes the task of the inspection point to be inspected;

[0054]

[0055] Where floor means rounding down, i means the number of the inspection point, s iIndicates the number of robots that can be selected at the position where the inspection point number to be inspected is i, z i ∈[1,i] indicates the serial number of the robot assigned to the inspection point i in the set of robots that can be selected at this inspection point, that is, the z i th element in the set of robots that can be assigned to the inspection point i is the robot number for performing the task of this inspection point;

[0056] The conversion of the coding form to the order of the robots performing the inspection tasks. For each inspection point assigned with a robot number obtained from the above formula (2), the first-layer elements of the inspection points assigned with robots with the same number are sorted in ascending order, and the order of the robots performing the tasks of each inspection point is represented by the sorting size;

[0057] Step 3: After establishing the mapping relationship between the coding method and the solution space by the two-layer coding method, use the improved hybrid artemisinin optimization algorithm to allocate appropriate inspection tasks for the robots and optimize the inspection route;

[0058] In the MATLAB simulation software, use the improved hybrid artemisinin optimization algorithm to solve the optimal route for each robot to start from the warehouse location, perform all the assigned inspection tasks, and finally return to the robot warehouse. The specific steps of the optimization algorithm are as follows:

[0059] 1) Introduce the Sigmoid function to improve the drug concentration factor C in the overall clearance stage of the artemisinin optimization algorithm, so as to improve the global search ability of the artemisinin optimization algorithm. Simulate the initial stage of artemisinin drug treatment for malaria, introduce the probability factor K. When rand < K, the movement position of the artemisinin drug particle is described as:

[0060]

[0061] K = 1 - t 1 / 6 / t max 1 / 6 (4)

[0062]

[0063] In the formula, and are the artemisinin positions before and after update respectively, is the current optimal position of artemisinin, t and t max are the current iteration number and the maximum iteration number respectively, and c represents the drug concentration factor;

[0064] 2) To further balance the global search ability and local search ability of the artemisinin optimization algorithm, in the local clearance stage, introduce the walking optimization algorithm to update the movement position of the artemisinin drug particle and improve the local search ability of the artemisinin algorithm. The walking optimization algorithm is described as follows:

[0065] s=tan(theta) (6)

[0066] vel=6e -3.5(s+0.05) (7)

[0067]

[0068] Where theta represents a random number in the set [0,1,2…50], which is used to simulate the trajectory or terrain, SF represents any number between 1 and 2, s represents the inclination angle of the trajectory or terrain, and vel represents the speed of the hiker;

[0069] 3) Late consolidation stage: In the late consolidation stage, patients are still likely to have malaria relapse; therefore, the movement position of artemisinin drug particles in the late consolidation stage is updated as follows:

[0070]

[0071] 4) Repeat the improvement of the mixed artemisinin optimization algorithm until the maximum number of iterations is completed, and select the best individual as the optimal route for the current robot to complete the assigned inspection task;

[0072] Specifically, MATLAB simulation software is used for calculation. Using a two-layer coding method, the improved hybrid artemisinin optimization algorithm solves the shortest path for each robot to start from the warehouse location to perform all assigned inspection tasks and finally return to the robot warehouse. The convergence curve is shown in Figure 3 As shown;

[0073] Step 4: Use MATLAB simulation software to visualize the distribution of robot inspection tasks in the chemical park and the results of path optimization. Figure 4 As shown;

[0074] Using MATLAB simulation software, the inspection points assigned to each robot obtained from steps one to three, as well as the optimal path for each robot to start from the warehouse, perform all assigned inspection tasks and finally return to the robot warehouse are visualized, making it convenient for actual chemical company staff to use.

[0075] Specifically, MATLAB simulation software is used to visualize the inspection points assigned to each robot obtained from steps 1 to 3, as well as the optimal path for each robot to start from the warehouse, complete all assigned inspection tasks, and finally return to the robot warehouse. Figure 4As shown in the figure, there are three robots R1, R2, and R3 in a chemical park to perform inspection tasks, among which R1 is a quadruped robot, R2 and R3 are wheeled robots, the inspection route of robot R1 is represented by a solid line, and the inspection routes of robots R2 and R3 are represented by a dotted line and a dotted line respectively. Circles and triangles are used to represent inspection points. The inspection points represented by circles can only be reached and completed by the R1 quadruped robot with stronger passing ability due to the terrain environment. The inspection points represented by triangles can be reached and completed by all three robots. The black part represents obstacles, and the five-pointed star in the lower left corner represents the robot's warehouse, that is, the starting point and end point of the robot.

[0076] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An intelligent method for multi-heterogeneous robot inspection task allocation and path optimization for chemical parks, characterized by: It includes the following steps: Step 1: Extract the environmental information of the chemical industrial park; Step 2: Establish the mapping relationship between the coding method and the solution space by setting a two-layer coding method; Step 3: After establishing the mapping relationship between the coding method and the solution space by the two-layer coding method, use the improved hybrid artemisinin optimization algorithm to allocate appropriate inspection tasks for the robot and optimize the inspection route; Step 4: Use MATLAB simulation software to visualize the information such as the allocation of the robot inspection tasks in the chemical industrial park and the results of the path optimization.

2. The intelligent method for multi-heterogeneous robot inspection task allocation and path optimization for chemical parks according to claim 1 is characterized in that: The specific content of Step 1 is as follows: The environmental information of the chemical industrial park includes the positions of the inspection points, the requirements of the inspection points for the robot inspection ability, the position of the robot warehouse, the routes in the chemical industrial park, and obstacles and other information, and store the environmental information of the chemical industrial park in an Excel table; Then use MATLAB simulation software to read the table data, using A * The algorithm calculates the route trajectory and journey cost between each inspection point, providing data for subsequent robot inspection task allocation and route optimization.

3. The intelligent method for multi-heterogeneous robot inspection task allocation and path optimization for chemical parks according to claim 1 is characterized in that: The specific content of Step 2 is as follows: After reading the Excel table storing the environmental information of the chemical industrial park with MATLAB simulation software, establish the mapping relationship between the coding method and the solution space of the inspection tasks assigned to the robot and the execution order of the inspection tasks by setting a two-layer coding method. This problem includes two sub-problems: allocating the inspection points to be inspected for the robot and sorting the order of the robot to execute the tasks of the inspection points to be inspected. Adopt a two-layer coding based on random keys for each individual, where the length of each layer is equal, and its coding form can be described as: Among them, the first layer is used to represent the priority order of the inspection points to be inspected by the robot, and the second layer is used to represent the robot number assigned to the inspection points to be inspected; each element in X takes any value within [-ε, ε]; the value of ε can take any appropriate integer, and l represents the number of inspection points to be inspected in the chemical industrial park; Map the encoding in formula (1) to the solution space of allocating inspection tasks for the robot and the order of the robot to execute the inspection tasks. According to formula (2), convert the element values of the individual position vector into the serial numbers in the set of robots that can be selected for the inspection positions to be inspected, and thus the robot number for executing the tasks of the inspection points to be inspected can be obtained; Where floor means rounding down, i means the number of the inspection point, s i represents the number of robots that can be selected at the position of the inspection point number i, z i ∈[1,s i ] represents the sequence number of the robot assigned to the inspection point i in the set of optional robots for the inspection point, that is, the robot that can be assigned to the inspection point i is the zth robot in the set of optional robots for the inspection point i The first element is the robot number that performs the task of the inspection point; The conversion of the coding form to the order of the robot to execute the inspection tasks. For each inspection point assigned with the same robot number obtained from the above formula (2), sort the first-layer elements of the inspection points assigned to the robots in ascending order, and use the size of the sorting to represent the order of the robot to execute the tasks of each inspection point.

4. The intelligent method for multi-heterogeneous robot inspection task allocation and path optimization for chemical parks according to claim 1 is characterized in that: The specific content of Step 3 is as follows: In the MATLAB simulation software, use the improved hybrid artemisinin optimization algorithm to solve the optimal route for each robot starting from the warehouse position, executing all the assigned inspection tasks, and finally returning to the robot warehouse; 1) Introduce the Sigmoid function in the comprehensive clearance stage of the artemisinin optimization algorithm to improve the drug concentration factor C, thereby improving the global search ability of the artemisinin optimization algorithm. Simulate the initial stage of artemisinin drug treatment for malaria, and introduce the probability factor K. When rand < k, the movement position of the artemisinin drug particle is described as: k=1-t 1 / 6 / t max 1 / 6 (4) Where: and are the positions of artemisinin before and after the update, is the current optimal position of artemisinin, t, t max are the current iteration number and the maximum iteration number, respectively, and c represents the drug concentration factor; 2) To further balance the global search ability and local search ability of the artemisinin optimization algorithm, in the local clearance stage, introduce the walking optimization algorithm to update the movement position of the artemisinin drug particle and improve the local search ability of the artemisinin algorithm. The walking optimization algorithm is described as follows: s = tan(theta) (6) vel=6e -3.5(s+0.05) (7) Where theta represents a random number in the set [0,1,2…50], which is used to simulate the trajectory or terrain, SF represents any number between 1 and 2, s represents the inclination angle of the trajectory or terrain, and vel represents the speed of the hiker; 3) Late consolidation stage: In the late consolidation stage, patients are still likely to have relapses of malaria; therefore, the movement position of artemisinin drug particles in the late consolidation stage is updated as follows: 4) Repeatedly improve the mixed artemisinin optimization algorithm until the maximum number of iterations is completed, and select the best individual as the optimal task allocation and inspection route plan for the robot.

5. The intelligent method for multi-heterogeneous robot inspection task allocation and path optimization for chemical parks according to claim 1 is characterized in that: The step four is specifically as follows: using MATLAB simulation software, the inspection points assigned to each robot obtained by solving the steps one to three, as well as the optimal path for each robot to start from the warehouse, perform all assigned inspection tasks and finally return to the robot warehouse are visualized, thereby facilitating use by actual chemical enterprise staff.

Images