Nuclear emergency robot path planning method based on improved ant colony algorithm

Abstract

The invention discloses a nuclear emergency robot path planning method based on an improved ant colony algorithm, and the method comprises the steps: firstly combining the ant colony algorithm with a Dijkstra algorithm, enhancing the initial pheromone concentration, and improving the convergence speed of the ant colony algorithm; on the basis, a dynamic segmented search method is designed, and a path is adjusted according to radiation distribution data received by the robot in the moving process in combination with an improved ant colony algorithm, so that the robot can avoid a high-radiation area. Besides, the energy consumption and the accumulated radiation dose are added into the heuristic function of the state transition probability formula to serve as heuristic information, the accumulated radiation dose on the path and the energy consumption of the robot are reduced, and the task efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of path planning, and particularly relates to the design of a path planning method for a nuclear emergency robot based on an improved ant colony algorithm. Background Technique

[0002] Nuclear emergency handling is the field where robots have the strongest demand and the widest application in nuclear environment operations. In case of an emergency, it is necessary to handle the opening and closing of key valves, repair key equipment, remove roadblocks, and radioactive substances with leakage, and path planning technology is the key to ensuring the efficient and safe execution of these tasks. Different from ordinary environments, path planning in a radiation environment not only needs to ensure the smooth completion of tasks, but also needs to minimize the dose of radiation to which the robot is exposed, so as to ensure the safety, optimality, and real-time nature of the path. Therefore, in the complex environment of nuclear emergency handling, reasonable path planning can not only effectively reduce the cumulative radiation dose of the robot in the radiation field, but also improve the operation efficiency, ensure that the robot completes the established tasks safely and efficiently, and provide guarantee for nuclear emergency response.

[0003] Path planning methods are generally divided into path planning methods based on graph search, path planning methods based on sampling, and heuristic optimization methods, etc. Graph search algorithms represented by the A* algorithm, although they can effectively plan the shortest path in a static environment, the paths generated by them may contain more inflection points, which not only affects the path smoothness in practical applications, but also leads to a higher computational complexity. Random sampling algorithms represented by the RRT algorithm, although they can effectively handle high-dimensional complex environments, due to the randomness of the algorithm, the paths generated by them are often unstable, and only feasible solutions can be obtained, and the optimal solution cannot be guaranteed to be found. Heuristic optimization methods such as genetic algorithms and particle swarm optimization algorithms, although they can search for better paths through optimization mechanisms, the convergence speeds of these algorithms are slower. When the search space is complex, it is difficult for these algorithms to guarantee finding the global optimal solution, and when dealing with large-scale problems, the consumption of computing resources is large, resulting in low efficiency. Most of the above path planning algorithms are applicable to simple nuclear radiation environments. In addition, because their implementation is complex and the computational cost is high, they are not suitable for application in emergency situations.

[0004] However, as a typical intelligent optimization algorithm, the ant colony algorithm has excellent parallel computing characteristics and can achieve efficient path planning under various constraints. Its pheromone update mechanism makes the computational consumption gradually decrease, thereby effectively reducing the energy consumption of the robot. And the ant colony algorithm has a strong global search ability and can effectively explore the optimal path in a complex dynamic environment. At the same time, by combining local search with global information, it can effectively avoid the problem of falling into local optimal solutions. Summary of the Invention

[0005] The object of the present invention is to solve the technical problems that existing path planning methods are difficult to obtain the path with the minimum cumulative radiation dose in an environment where the radiation dose distribution changes dynamically, the algorithm convergence speed is low, and the robot has high energy consumption. A path planning method for a nuclear emergency robot based on an improved ant colony algorithm is proposed, which effectively realizes efficient and safe path planning for a nuclear emergency robot in a complex dynamic radiation dose environment.

[0006] The technical solution of the present invention is as follows: A path planning method for a nuclear emergency robot based on an improved ant colony algorithm includes the following steps:

[0007] S1. Through the improved ant colony algorithm, an initial path from the starting point to the target point is planned at the initial moment.

[0008] S2. The initial path is divided into n segments, and the currently planned path segment number h = 1 is set.

[0009] S3. Control the nuclear emergency robot to start moving along the first divided path segment.

[0010] S4. In response to the nuclear emergency robot reaching the end point of the first path segment, take the current position of the nuclear emergency robot as the new starting point, keep the target point unchanged, and plan a current path from the new starting point to the target point through the improved ant colony algorithm.

[0011] S5. The current path is divided into n - h segments.

[0012] S6. Judge whether n - h = 1 is satisfied. If so, enter step S7; otherwise, increase the value of h by 1 and return to step S3.

[0013] S7. Control the nuclear emergency robot to move along the current path to the target point.

[0014] Furthermore, the improved ant colony algorithm includes the following steps:

[0015] A1. Improve the pheromone concentration at the initial moment of the ant colony algorithm through the Dijkstra algorithm.

[0016] A2. Initialize and set the maximum number of ants and the maximum number of iterations of the ant colony algorithm, and set the current iteration number t = 1.

[0017] A3. Set the current ant serial number k = 1.

[0018] A4. Obtain the starting point and the target point.

[0019] A5. Select the next moving direction of ant k through the state transition probability formula.

[0020] A6. Judge whether ant k has reached the target point. If so, enter step A7; otherwise, return to step A5.

[0021] A7. Determine whether k has reached the maximum number of ants. If so, go to step A8; otherwise, increment the ant sequence number k by 1 and return to step A4.

[0022] A8. Record the path in the current iteration and update the pheromone concentration on the path.

[0023] A9. Determine whether t has reached the maximum number of iterations. If so, go to step A10; otherwise, increment the current iteration number t by 1 and return to step A3.

[0024] A10. Compare the paths of each iteration, select the path with the minimum cumulative radiation dose as the optimal path and output it.

[0025] Further, the formula for improving the pheromone concentration at the initial moment of the ant colony algorithm in step A1 by the Dijkstra algorithm is:

[0026]

[0027] where τ ij ′(0) represents the pheromone concentration at the initial moment of the ant colony algorithm before improvement, and τ ij (0) represents the pheromone concentration at the initial moment of the ant colony algorithm after improvement, and C represents the enhancement value of the pheromone concentration at the initial moment.

[0028] Further, the state transition probability formula in step A5 is:

[0029]

[0030] where represents the state transition probability that ant k moves from the current node i to the next node j at the t-th iteration, τ ij (t) represents the pheromone concentration on the path from the current node i to the next node j at the t-th iteration, η ij (t) represents the heuristic function from the current node i to the next node j at the t-th iteration, α represents the pheromone influence weight coefficient, β represents the heuristic function influence weight coefficient, s represents the set of optional grids around ant k, and allowed k represents the set of the next optional nodes.

[0031] Further, the formula for the heuristic function η ij (t) is:

[0032]

[0033] where D ij represents the cumulative radiation dose between the current node i and the next node j, and E ijIt represents the energy consumption between the current node i and the next node j, and Q represents the enhancement coefficient.

[0034] Furthermore, the energy consumption E ij is calculated by the formula:

[0035] E ij = μ1angle + μ2turn

[0036] angle = |θ c_l - θ n_c |

[0037] θ n_c = atan2(P next , P current )

[0038] θ c_l = atan2(P current , P last )

[0039]

[0040] where μ1 represents the coefficient for converting the turning angle variable to the grid length, μ2 represents the weight coefficient for the number of turns, angle represents the turning angle, turn represents the number of turns, θ c_l represents the angle between the straight line l formed by the previous node and the current node c_l and the horizontal straight line, θ n_c represents the angle between the straight line l formed by the current node and the next node n_c and the horizontal straight line, P current represents the current node, P last represents the previous node, P next represents the next node, and atan2(·) represents the arctangent between two points.

[0041] Furthermore, the formula for updating the pheromone concentration on the path in step A8 is:

[0042] τ ij (t + 1) = (1 - ρ)τ ij (t) + ρ′Δτ ij (t)

[0043]

[0044] where τ ij (t + 1) represents the pheromone concentration on the path from the current node i to the next node j at the (t + 1)-th iteration, ρ represents the pheromone evaporation factor, ρ′ represents the pheromone increase factor, and Δτ ij (t) represents the pheromone increment left by all ants on the path from the current node i to the next node j at the t-th iteration. It represents the pheromone increment left by ant k on the path from the current node i to the next node j at the t-th iteration. m represents the maximum number of ants, Q1 represents the pheromone increase intensity coefficient of the cumulative radiation dose, Q2 represents the pheromone increase intensity coefficient of the energy consumption, and D min represents the minimum radiation dose in this iteration, and E min represents the minimum energy consumption in this iteration.

[0045] Furthermore, when updating the pheromone concentration on the path in step A8, the following constraints are satisfied:

[0046]

[0047] where τ ij (t) min represents the minimum pheromone concentration, and τ ij (t) max represents the maximum pheromone concentration.

[0048] The beneficial effects of the present invention are as follows:

[0049] (1) By designing the dynamic segmented search method, the present invention effectively solves the problem of solving the minimum dose in an environment with dynamic radiation dose changes. In this way, the nuclear emergency robot can continuously adjust the path in the dynamic radiation dose environment, thereby reducing the cumulative radiation dose of the planned path.

[0050] (2) By combining the global optimization ability of the Dijkstra algorithm, the present invention improves the initial pheromone distribution of the ant colony algorithm, thereby improving the convergence speed of the ant colony algorithm. This improvement indirectly reduces the time required for path search and effectively reduces the radiation dose on the path.

[0051] (3) By optimizing the pheromone update rule in the ant colony algorithm, the present invention overcomes the possible search stagnation phenomenon in the traditional ant colony algorithm and avoids the algorithm falling into a local optimal solution during the search process.

[0052] (4) The present invention takes the angular variable and the number of turns of the path as the main influencing factors of the energy consumption index. The angular variable and the number of turns in the path reflect the curvature and complexity of the robot's trajectory. These factors not only directly affect the driving distance and energy consumption of the robot, but also indirectly affect the cumulative radiation dose of the robot during operation. By adding energy consumption as heuristic information to the heuristic function of the ant colony algorithm, the present invention reduces the energy consumption on the route, thereby indirectly reducing the cumulative radiation dose of the robot during operation. Description of the Drawings

[0053] Figure 1The figure shows a flowchart of a path planning method for a nuclear emergency robot based on an improved ant colony algorithm provided by an embodiment of the present invention.

[0054] Figure 2 The figure shows a flowchart of the improved ant colony algorithm provided by an embodiment of the present invention.

[0055] Figure 3 The figure shows a schematic diagram of the path angle provided by an embodiment of the present invention.

[0056] Figure 4 The figure shows a schematic diagram of the dynamic segmentation search method provided by an embodiment of the present invention. Detailed implementation manners

[0057] Now, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be understood that the embodiments shown and described in the drawings are merely exemplary, intended to illustrate the principles and spirit of the present invention, and not to limit the scope of the present invention.

[0058] An embodiment of the present invention provides a path planning method for a nuclear emergency robot based on an improved ant colony algorithm. As Figure 1 shown, it includes the following steps S1 to S7:

[0059] S1. At the initial moment, plan an initial path from the starting point to the target point through the improved ant colony algorithm.

[0060] S2. Divide the initial path into n segments, and set the currently planned number of path segments h = 1.

[0061] S3. Control the nuclear emergency robot to start moving along the first divided path segment.

[0062] S4. In response to the nuclear emergency robot reaching the end point of the first path segment, use the current position of the nuclear emergency robot as the new starting point, keep the target point unchanged, and plan a current path from the new starting point to the target point through the improved ant colony algorithm.

[0063] S5. Divide the current path into n - h segments.

[0064] S6. Determine whether n - h = 1 is satisfied. If so, enter step S7; otherwise, increment the value of h by 1 and return to step S3.

[0065] S7. Control the nuclear emergency robot to move along the current path to the target point.

[0066] The ant colony algorithm is a heuristic optimization algorithm that simulates the foraging behavior of ants. Its basic idea is to guide the behavior of other ants by simulating the pheromones released by ants during the process of finding food. Ants leave pheromones on the path and choose the best path according to the concentration of pheromones on the path. As the iteration progresses, the concentration of pheromones on the shorter path is higher, attracting more ants to choose this path, thus finding the optimal solution to the problem.

[0067] In the embodiments of the present invention, by optimizing the pheromone update rule in the traditional ant colony algorithm, the phenomenon of search stagnation that may occur in the traditional ant colony algorithm is overcome, and the algorithm is prevented from falling into a local optimal solution during the search process. The improved ant colony algorithm is applied to path planning in steps S1 and S4, and specifically includes the following steps A1 to A10:

[0068] A1. Improve the pheromone concentration at the initial moment of the ant colony algorithm through the Dijkstra algorithm.

[0069] In the embodiments of the present invention, first, in order to reduce the early search time of the algorithm and improve the search efficiency, the initial pheromone distribution is improved. When initializing the pheromone, first use the Dijkstra algorithm to generate a path K, increase the pheromone concentration on path K, and keep the pheromone concentration on other paths unchanged. The specific formula is:

[0070]

[0071] where τ ij ′(0) represents the pheromone concentration at the initial moment of the ant colony algorithm before improvement, τ ij (0) represents the pheromone concentration at the initial moment of the ant colony algorithm after improvement, C represents the enhancement value of the pheromone concentration at the initial moment, and is a constant.

[0072] A2. Initialize and set the maximum number of ants and the maximum number of iterations of the ant colony algorithm, and set the current iteration number t = 1.

[0073] A3. Set the current ant serial number k = 1.

[0074] A4. Obtain the starting point and the target point.

[0075] A5. Select the next moving direction of ant k through the state transition probability formula.

[0076] In the ant colony algorithm, the next moving direction of an ant is determined by the state transition probability, and the state transition probability is mainly determined by the pheromones on the path and the heuristic function. In the embodiments of the present invention, the state transition probability formula is:

[0077]

[0078] Among them represents the state transition probability that ant k moves from the current node i to the next node j at the t-th iteration, τ ij (t) represents the pheromone concentration on the path from the current node i to the next node j at the t-th iteration, η ij (t) represents the heuristic function from the current node i to the next node j at the t-th iteration, α represents the pheromone influence weight coefficient, β represents the heuristic function influence weight coefficient, s represents the set of optional grids around ant k, allowed k represents the set of the next optional nodes.

[0079] In the embodiment of the present invention, to ensure that the nuclear emergency robot obtains a path with a small cumulative radiation dose and low energy consumption during path planning, the energy consumption and the cumulative radiation dose are added as heuristic information in the heuristic function η ij (t), specifically:

[0080]

[0081] where D ij represents the cumulative radiation dose between the current node i and the next node j, E ij represents the energy consumption between the current node i and the next node j, and Q represents the enhancement coefficient.

[0082] In the embodiment of the present invention, the energy consumption E ij is an index used to quantify the comprehensive energy consumption of the path, and evaluates the overall influence of the cumulative value of the turning angles and the number of turns on the energy consumption. Its specific calculation formula is:

[0083] E ij =μ1angle + μ2turn

[0084] angle = |θ c_l -θ n_c |

[0085] θ n_c =atan2(P next ,P current )

[0086] θ c_l =atan2(P current ,P last )

[0087]

[0088] where μ1 represents the coefficient for converting the turning angle variable to the grid length, μ2 represents the weight coefficient of the number of turns, angle represents the turning angle, turn represents the number of turns, θ c_lDenote the straight line \(l\) formed by the previous node and the current node c_l and the included angle with the horizontal straight line, \(\theta\) n_c Denote the straight line \(l\) formed by the current node and the next node n_c and the included angle with the horizontal straight line, \(\varphi\) current Denote the current node, \(P\) last Denote the previous node, \(P\) next Denote the next node, and \(\text{atan2}(\cdot)\) represents the arctangent between two points.

[0089] The schematic diagram of the path included angle is as Figure 3 shown Figure 3 in which \(\text{angle}(l\) n_c ,l c_l ) represents the path included angle, that is, the turning angle, \(P\) last (x last ,y last ) is the previous node on the current path, \(P\) current (x current ,y current ) is the current node on the current path, \(P\) next (x next ,y next ) is the next node on the current path.

[0090] A6. Determine whether ant \(k\) reaches the target point. If so, go to step A7; otherwise, return to step A5.

[0091] A7. Determine whether \(k\) reaches the maximum number of ants. If so, go to step A8; otherwise, increment the ant serial number \(k\) by 1 and return to step A4.

[0092] A8. Record the path in the current iteration and update the pheromone concentration on the path.

[0093] When each ant searches for a path, it releases pheromone with a certain volatility, resulting in the continuous accumulation and volatilization of the total pheromone amount on each path. To overcome the possible stagnation phenomenon in the existing ant colony system, the embodiment of the present invention introduces the max - min ant system to improve the update of the pheromone concentration. The specific formula is as follows:

[0094] \(\tau\) ij (t + 1)=(1-\rho)\tau ij (t)+\rho'\Delta\tau ij (t)

[0095]

[0096] where \(\tau\) ij(t + 1) represents the pheromone concentration on the path from the current node i to the next node j at the (t + 1)-th iteration, ρ represents the pheromone evaporation factor, ρ′ represents the pheromone increase factor, and Δτ ij (t) represents the increment of pheromone left by all ants on the path from the current node i to the next node j at the t-th iteration. represents the increment of pheromone left by ant k on the path from the current node i to the next node j at the t-th iteration, m represents the maximum number of ants, Q1 represents the pheromone increase intensity coefficient of the cumulative radiation dose, Q2 represents the pheromone increase intensity coefficient of the energy consumption, and D min represents the minimum radiation dose in this iteration, and E min represents the minimum energy consumption in this iteration.

[0097] Meanwhile, in order to avoid falling into local optimum during the search process, the pheromone concentration is generally restricted within [τ min , τ max , which ensures that all paths in the search space have a certain chance to be selected instead of only choosing the paths with high pheromone concentration. Therefore, when updating the pheromone concentration on the path in the embodiments of the present invention, the following constraints are satisfied:

[0098]

[0099] where τ ij (t) min represents the minimum pheromone concentration, and τ ij (t) max represents the maximum pheromone concentration.

[0100] A9. Determine whether t reaches the maximum number of iterations. If so, go to step A10; otherwise, increment the current iteration number t by 1 and return to step A3.

[0101] A10. Compare the paths of each iteration, and select the path with the minimum cumulative radiation dose as the optimal path and output it.

[0102] The dynamic segmented search method adopted in the embodiments of the present invention is as Figure 4 shown. It can be seen from Figure 4 that through the dynamic segmented search method, the path can be adjusted according to the radiation distribution data received by the robot during movement in combination with the improved ant colony algorithm, enabling the nuclear emergency robot to avoid high-radiation areas.

[0103] Those of ordinary skill in the art will realize that the embodiments described herein are for the purpose of assisting the reader in understanding the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific deformations and combinations that do not depart from the essence of the present invention based on these technical revelations disclosed in the present invention, and these deformations and combinations are still within the scope of protection of the present invention.

Claims

1. A path planning method for nuclear emergency robots based on an improved ant colony algorithm, characterized in that It includes the following steps: S1. Plan an initial path from the starting point to the target point at the initial moment by improving the ant colony algorithm; S2. Divide the initial path into n segments, and set the currently planned number of path segments h = 1; S3. Control the nuclear emergency robot to start moving along the first divided path segment; S4. In response to the nuclear emergency robot reaching the end point of the first path segment, take the current position of the nuclear emergency robot as the new starting point, keep the target point unchanged, and plan a current path from the new starting point to the target point by improving the ant colony algorithm; S5. Divide the current path into n - h segments; S6. Judge whether n - h = 1 is satisfied. If so, enter step S7; otherwise, increase the value of h by 1 and return to step S3; S7. Control the nuclear emergency robot to move along the current path to the target point.

2. The method for path planning of a nuclear emergency robot based on an improved ant colony algorithm according to claim 1, characterized in that The improved ant colony algorithm includes the following steps: A1. Improve the pheromone concentration at the initial moment of the ant colony algorithm by the Dijkstra algorithm; A2. Initialize and set the maximum number of ants and the maximum number of iterations of the ant colony algorithm, and set the current number of iterations t = 1; A3. Set the current ant serial number k = 1; A4. Obtain the starting point and the target point; A5. Select the next moving direction of ant k through the state transition probability formula; A6. Judge whether ant k reaches the target point. If so, enter step A7; otherwise, return to step A5; A7. Judge whether k reaches the maximum number of ants. If so, enter step A8; otherwise, increase the ant serial number k by 1 and return to step A4; A8. Record the path in the current iteration and update the pheromone concentration on the path; A9. Judge whether t reaches the maximum number of iterations. If so, enter step A10; otherwise, increase the current number of iterations t by 1 and return to step A3; A10. Compare the paths of each iteration, select the path with the minimum cumulative radiation dose as the optimal path and output it.

3. The method for path planning of a nuclear emergency robot based on an improved ant colony algorithm according to claim 2, characterized in that, The formula for improving the pheromone concentration at the initial moment of the ant colony algorithm by the Dijkstra algorithm in step A1 is: where τ ij '(0) represents the pheromone concentration at the initial moment of the ant colony algorithm before improvement, and τ ij (0) represents the pheromone concentration at the initial moment of the improved ant colony algorithm, and C represents the enhancement value of the pheromone concentration at the initial moment.

4. The method for path planning of a nuclear emergency robot based on an improved ant colony algorithm according to claim 2, wherein The state transition probability formula in step A5 is: Among them represents the state transition probability that ant k moves from the current node i to the next node j at the t-th iteration, τ ij (t) represents the pheromone concentration on the path from the current node i to the next node j at the t-th iteration, η ij (t) represents the heuristic function from the current node i to the next node j at the t-th iteration, α represents the pheromone influence weight coefficient, β represents the heuristic function influence weight coefficient, s represents the set of optional grids around ant k, allowed k represents the set of the next optional nodes.

5. The method for path planning of a nuclear emergency robot based on an improved ant colony algorithm according to claim 4, wherein The heuristic function η ij (t) is given by the formula: Among which D ij represents the cumulative radiation dose between the current node i and the next node j, and E ij represents the energy consumption between the current node i and the next node j, and Q represents the enhancement coefficient.

6. The method for path planning of a nuclear emergency robot based on an improved ant colony algorithm according to claim 5, wherein, The energy consumption E ij has the following calculation formula: E ij = μ1angle + μ2turn angle = |θ c_l -θ n_c | θ n_c = atan2(P next , P current ) θ c_l = atan2(P current , P last ) Among them, μ1 represents the coefficient for converting the turning angle variable to the grid length, μ2 represents the weight coefficient of the number of turns, angle represents the turning angle, turn represents the number of turns, and θ c_l represents the angle between the straight line l c_l formed by the previous node and the current node and the horizontal straight line, and θ n_c represents the angle between the straight line l n_c formed by the current node and the next node and the horizontal straight line, and P current represents the current node, and P last represents the previous node, and P next represents the next node, and atan2(·) represents the arctangent between two points.

7. The path planning method of the nuclear emergency robot based on the improved ant colony algorithm according to claim 2, wherein The formula for updating the pheromone concentration on the path in step A8 is: τ ij (t + 1) = (1 - ρ)τ ij (t) + ρ′Δτ ij (t) where τ ij (t + 1) represents the pheromone concentration on the path from the current node i to the next node j at the (t + 1)-th iteration, ρ represents the pheromone evaporation factor, ρ′ represents the pheromone increase factor, and Δτ ij (t) represents the pheromone increment left by all ants on the path from the current node i to the next node j at the t-th iteration, represents the pheromone increment left by ant k on the path from the current node i to the next node j at the t-th iteration, m represents the maximum number of ants, Q1 represents the cumulative radiation dose pheromone increase intensity coefficient, Q2 represents the energy consumption pheromone increase intensity coefficient, D min represents the minimum radiation dose in this iteration, and E min represents the minimum energy consumption in this iteration.

8. The method for path planning of nuclear emergency robots based on the improved ant colony algorithm according to claim 7, characterized in that When updating the pheromone concentration on the path in step A8, the following constraints are satisfied: where τ ij (t) min represents the minimum pheromone concentration, and τ ij (t) max represents the maximum pheromone concentration.

Images