Mobile robot indoor kernel environment multi-target point inspection path planning method

Abstract

The mobile robot indoor kernel environment multi-target point inspection path planning method, the grid map of the indoor kernel environment is prior information, each grid in the grid map contains radiation field information and obstacle information; the method is as follows: 1. Multi-target point inspection path mathematical modeling: considering path length and cumulative radiation dose and turning energy consumption, respectively, an inspection path evaluation model and a pair of path evaluation model are established; 2, the upper algorithm is used to obtain the inspection path; 3, the lower algorithm is used to calculate the pair of path; 4, the optimal solution is obtained by the circulation of the two-layer algorithm. The advantages of the present application are: for the mobile robot MTIPP problem in the indoor kernel environment, the path length, the cumulative radiation dose and the turning energy consumption are considered, a double-layer mathematical model and a double-layer hybrid algorithm are established, and the inspection path obtained by planning has better safety and lower global cost compared with the path planning considering only single factor in the prior art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of path planning, in particular to a mobile robot indoor nuclear environment multi-target point inspection path planning method. BACKGROUND

[0002] Multi-target point inspection tasks often need to be performed in the indoor environment of a nuclear power plant. Under the premise that the indoor map of the nuclear power plant (which contains radiation field and obstacle information) is known, planning a suitable inspection path is the basis for the safe and smooth execution of multi-target point inspection tasks. The above-mentioned multi-target point inspection tasks are usually performed by manual inspection. The path planning of manual inspection mostly only considers the cumulative radiation dose, and the path planning difficulty is relatively low.

[0003] It is rare to see mobile robots performing multi-target point inspection tasks. The path planning of multi-target point inspection tasks is difficult, and it is necessary to consider the turning energy consumption of the mobile robot (as few turns as possible and as small a turning angle as possible), the cumulative radiation dose (as low as possible under the premise that it does not exceed the bearing limit of the mobile robot), and the path length (the shorter the better). Currently, there is no path planning method for mobile robots to perform multi-target point inspection tasks in the industry. SUMMARY

[0004] The purpose of the present application is to overcome the shortcomings of the prior art and provide an indoor nuclear environment multi-target point inspection path planning method, which solves the problem of how to reasonably plan an inspection path for a mobile robot multi-target point inspection task under the premise that the indoor nuclear environment map is known.

[0005] The technical solution of the present application is: a mobile robot indoor nuclear environment multi-target point inspection path planning method, the grid map of the indoor nuclear environment is prior information, and each grid in the grid map contains radiation field information and obstacle information;

[0006] The method is as follows:

[0007] S01, multi-target point inspection path mathematical modeling: considering the path length, cumulative radiation dose and turning energy consumption, respectively establishing an inspection path advantage and disadvantage evaluation model and a pair of path advantage and disadvantage evaluation model;

[0008] S02, obtaining an inspection path based on an upper algorithm: using a GA algorithm to obtain a rough optimization solution of the inspection path; using the rough optimization solution of the inspection path to process the initial pheromone non-uniformity in the ACO algorithm; considering the path length, cumulative radiation dose and turning energy consumption, constructing the heuristic function and selection probability function of the ACO algorithm, and optimizing the pheromone update rule of the ACO algorithm; completing the construction of the upper algorithm, running the upper algorithm, and obtaining a fine optimization solution of the inspection path;

[0009] S03, constructing the lower layer algorithm to calculate the pair path: considering the path length and the accumulated radiation dose and the turning energy consumption, establishing the heuristic function of A* algorithm; completing the construction of the lower layer algorithm, running the lower layer algorithm, and planning all the pair paths in the fine optimization solution of the inspection path;

[0010] S04, the optimal solution of the two-layer algorithm loop running: using the inspection path evaluation model to compare the pros and cons of the fine optimization solution of the inspection path obtained by the current iteration and the last iteration, if the fine optimization solution of the inspection path obtained by the current iteration is better, the fine optimization solution of the inspection path obtained by the current iteration is put into the next iteration; if the fine optimization solution of the inspection path obtained by the current iteration is worse or consistent, the iteration is ended, and the fine optimization solution of the inspection path obtained by the last iteration is output.

[0011] The further technical solution of the application is as follows:

[0012] A, establishing the inspection path evaluation model: the multi-target point inspection task needs to establish a Hamilton circuit for each inspection target point to pass through only once and finally return to the starting point, considering the path length and the accumulated radiation dose and the turning energy consumption in three dimensions, establishing the inspection path evaluation model, or the inspection target point traversal order evaluation model, refer to formula 1;

[0013] Formula 1:

[0014] Wherein, λ is the total value of the three dimensions of all pair paths in the inspection path; T is the set of inspection target points, T={T1, T2, T3, …, Tn}, the subscript n is the total number of inspection target points; is the path length cost matrix between the inspection target points T i and T j , is the accumulated radiation dose cost matrix between the inspection target points T i and T j , is the turning energy consumption cost matrix between the inspection target points T i and T j ; τi ,j is a decision variable, if the value is 1, it indicates that the path between the inspection target points T i and T j is selected, if the value is 0, it indicates that the path between the inspection target points T i and T j is not selected; ω1 is the weight coefficient of the path length, ω2 is the weight coefficient of the accumulated radiation dose, and ω3 is the weight coefficient of the turning energy consumption;

[0015] B, establishing the pair path cost calculation model: T i and Tj are any two inspection target points, let T i be the path from T j to T m with the minimum total value of three dimensions, the grid set sequentially passed by the path is P, P = {p1, p, p3…p i}, then the grid where T j is located is p1, and the grid where T m is located is p i ; considering the path length, the cumulative radiation dose and the turning energy consumption of three dimensions, a pair of path cost calculation model is established, as shown in formula 2;

[0016] Formula 2:

[0017] Wherein, M(P) is the total value of three dimensions of the pair of path T j to T i , L(P) is the path length value of T j to T i , R(P) is the cumulative radiation dose value of T j to T i , E(P) is the turning energy consumption value of T j to T α-1 ; ε1 is the weight coefficient of the turning angle variable, ε2 is the weight coefficient of the turning number, l(pa-1, pa) is the center line of the grid p α and the grid p +1 , l(pa, pa α ) is the center line of the grid p α+1 and the grid p -1 , ψ[l(pa +1 ,pa),l(pa,pa -1 )] is the turning angle variable of the line segment l(pa +1 ,pa) and the line segment l(pa,pa α ), Nturn is the turning number; (xα,yα) is the coordinate of the grid p +1 , (xα +1 ,yα α+1 ) is the coordinate of the grid p α-1 , (xα-1,yα-1) is the coordinate of the grid p +1 ; R(pα,pα a ) is the cumulative radiation dose received by the mobile robot when moving from the grid p a+1 to the grid p +1 , is the average radiation dose rate of the grid pα, is the average radiation dose rate of the grid pα +1the distance between the grid p and p +1 is equal to 1 times the grid edge length, if the grid p and p +1 end point contact, is equal to times the grid edge length, v is the speed from the grid p to the grid p +1 .

[0018] The further technical solution of the present application is that the S02 step is described in detail as follows:

[0019] A, the GA algorithm is used to obtain the rough optimization solution of the inspection path:

[0020] The chromosome in the GA algorithm corresponds to the traversal order of the inspection target points, the gene in the GA algorithm corresponds to the inspection target points, and the fitness in the GA algorithm corresponds to the traversal order of the inspection target points. The evaluation function is as follows:

[0021] a1, real number coding: according to the numbering 1, 2, 3…n, the inspection target points are deployed in turn, indicating that the mobile robot should traverse n inspection target points, and a traversal order of the inspection target points is randomly generated as an initial population;

[0022] a2, create a taboo table: for each real gene coding, the selected taboo table is marked as 1, and the unselected taboo table is marked as 0;

[0023] a3, population initialization: set the evolution generation counter G and the maximum evolution generation G max , and randomly generate an initial population;

[0024] a4, individual evaluation: all the traversal orders of the inspection target points are evaluated according to the Hamilton circuit evaluation function;

[0025] a5, individual selection: select multiple traversal orders of the inspection target points with the smallest generation value as the operators, which are applied to the existing population, so that individuals with good genes are retained for the next generation;

[0026] a6, crossover operation: according to the preset crossover probability P c , select individuals from the existing population, and generate new offspring individuals through the crossover operator, thereby generating new traversal orders of the inspection target points;

[0027] a7, evolution termination judgment: when G≤G max , repeat a3-a6 steps; when G>G max , the evolution is terminated, and the individual with the smallest generation value in the population is the rough optimization solution of the inspection path, which is recorded as path GA .

[0028] B, the initial pheromone in the ACO algorithm is not uniform: the initial pheromone in the ACO algorithm is not uniform by using the coarse optimization solution of the inspection path obtained by the GA algorithm, as shown in formula 3;

[0029] Formula 3:

[0030] In the formula, τ 0 (i,j) represents the initial pheromone concentration after the non-uniform processing between the inspection target point T i and T j , the inspection target point T i and T j are the pair of paths, C is the initial pheromone weighting value, τ is the set initial pheromone concentration, and s.t. is the constraint condition;

[0031] C, the heuristic function of the optimized ACO algorithm: considering the influence of path length, cumulative radiation dose and turning energy consumption on the heuristic function, while introducing an adaptive adjustment coefficient to balance the convergence speed and global search ability of the ACO algorithm, the heuristic function of the ACO algorithm is constructed, as shown in formula 4;

[0032] Formula 4:

[0033] In the formula, η t (i,j) is the heuristic function value from the inspection target point T i to T j in the tth iteration; I c is the current iteration number, I max is the maximum iteration number; ξ is the adaptive adjustment coefficient; ω1 is the weight coefficient of the path length value, ω2 is the weight coefficient of the cumulative radiation dose value, ω3 is the weight coefficient of the turning energy consumption value, is the path length cost matrix from the inspection target point T i to T j , is the cumulative radiation dose cost matrix from the inspection target point T i to T j , is the turning energy consumption cost matrix from the inspection target point T i to T j ;

[0034] D, the transition probability function of the optimized ACO algorithm: considering the pheromone concentration and the heuristic function, the transition probability function of the ACO algorithm is constructed, as shown in formula 5;

[0035] Formula 5:

[0036] In the formula, P t(i,j) is the selection probability of the tth iteration from the inspection target point T i to T j ; α is the pheromone importance factor, β is the heuristic function importance factor; τ t (i,j) represents the pheromone concentration of the tth iteration from the inspection target point T i to T j ; τ t (i,u) represents the pheromone concentration of the tth iteration from the inspection target point T i to T u ; T t ' is the set of untraveled inspection target points in the tth iteration;

[0037] E, the pheromone update rule of the optimized ACO algorithm: considering the path length and the cumulative radiation dose and the turning energy consumption, the pheromone update rule of the ACO algorithm is optimized, as shown in formula 6; thus the construction of the upper algorithm is completed, and the optimized solution of the inspection path is obtained by running the upper algorithm;

[0038] Formula 6:

[0039] In the formula, τt +1 (i,j) represents the pheromone concentration of the t+1th iteration from the inspection target point T i to T j , τt(i,j) represents the pheromone concentration of the tth iteration from the inspection target point T i to T j , ρ is the pheromone evaporation factor, Q is the pheromone deposition factor, υ t-mi is the total three-dimensional value of the optimal inspection path in the tth iteration, υ t-ma is the total three-dimensional value of the worst inspection path in the tth iteration; I t is the optimal inspection path in the tth iteration; υ is the total three-dimensional value of the inspection path; assuming that there are n inspection target points in the inspection path, then “Ti,Ti +1 ” are two consecutive inspection target points in the inspection path, and “Tn,T1” are two consecutive inspection target points in the inspection path, and the inspection path requires each inspection target point to pass through only once and finally return to the starting point; is the path length cost matrix between the inspection target points T i and T i+1 , is the cumulative radiation dose cost matrix between the inspection target points T i and T i+1 , is the turning energy consumption cost matrix between the inspection target points T i and T i+1 , T is the inspection target point n a path length cost matrix between T and T1, T is the inspection target point n a cumulative radiation dose cost matrix between T and T1, T is the inspection target point n a turning energy consumption cost matrix between T and T1.

[0040] The further technical solution of the present application is that the S03 step is described in detail as follows:

[0041] Considering the path length, the cumulative radiation dose and the turning energy consumption, the heuristic function of the A* algorithm is established, all the paired paths in the inspection path fine optimization solution are planned by the A* algorithm, the starting point and the end point of the paired path are respectively named as the starting inspection target point and the end inspection target point; according to the three-dimensional cost values involved in the paired path merit evaluation model, the heuristic function of the A* algorithm is shown in formula 7; thus the construction of the lower algorithm is completed, the lower algorithm is run for multiple times, and the specific routes and the three-dimensional cost matrices of all the paired paths in the inspection path fine optimization solution are obtained;

[0042] Formula 7:

[0043] In the formula, (g(p)+h(p)) is equivalent to L(P) in formula 2, and is used to represent the path length cost value between two inspection target points; R(p 1-m ) is equivalent to R(P) in formula 2, and is used to represent the cumulative radiation dose cost value between two inspection target points; E(p 1-m ) is equivalent to E(P) in formula 2, and is used to represent the turning energy consumption cost value between two inspection target points; f(x) is the three-dimensional cost value between the starting inspection target point and the end inspection target point; g(p) is the path length cost value from the starting inspection target point to the current grid, h(p) is the path length cost value from the current grid to the end inspection target point, (x1,y1) is the coordinate of the grid p1 where the starting inspection target point is located, (x i ,y i ) is the coordinate of the current grid pi, and (x m ,y m ) is the coordinate of the grid pm where the end inspection target point is located.

[0044] The further technical solution of the present application is that the A* algorithm in the S03 step further includes a high radiation dose rate dangerous area avoidance strategy;

[0045] In the path search process, when the ant perceives that the radiation dose rate of the next grid exceeds the bearing limit of the mobile robot, the radiation dose rate of the next grid is defined as infinite, so as to prompt the ant to find other passable grids, and refer to formula 8.

[0046] Formula 8:

[0047] Wherein, P(pi) is the radiation dose rate of the grid pi, Inf is infinite, and RImax is the maximum radiation dose rate that can be borne by the mobile robot.

[0048] A further technical solution of the present application is that the S04 step is described in detail as follows:

[0049] Each time the "inspection path fine optimization solution" obtained by running the upper algorithm is marked as A1, A2, A3...Az in sequence; each time the "three-dimensional cost matrix of all pairs of paths in the inspection path fine optimization solution" obtained by running the lower algorithm is marked as C1, C2, C3...Cz in sequence; the results obtained by running the upper algorithm for the first time A1 and the results obtained by running the lower algorithm for the first time C1 are substituted into formula 1 to calculate the three-dimensional total cost value corresponding to A1, which is denoted as λ1.

[0050] A1 is taken as the initial population of the GA algorithm, the S02-S03 steps are repeatedly run to obtain A2 and C2; A2 and C2 are substituted into formula 1 to calculate the three-dimensional total cost value corresponding to A2, which is denoted as λ2; thus the first iteration is completed; λ1 and λ2 are compared, if λ2≥λ1, then A1 corresponding to λ1 is the global optimal solution of the inspection path, if λ2<λ1, then A2 is put into the next iteration;

[0051] A2 is taken as the initial population of the GA algorithm, the S02-S03 steps are repeatedly run to obtain A3 and C3; A3 and C3 are substituted into formula 1 to calculate the three-dimensional total cost value corresponding to A3, which is denoted as λ3; thus the second iteration is completed; λ2 and λ3 are compared, if λ3≥λ2, then A2 corresponding to λ2 is the global optimal solution of the inspection path, if λ3<λ2, then A3 is put into the next iteration;

[0052] In this way, the global optimal solution of the inspection path is finally obtained.

[0053] A further technical solution of the present application is that the grid in the grid map is a square; the radiation field information contained in each grid in the grid map is the average radiation dose rate recorded in each grid, and the obstacle information contained in each grid in the grid map is identified by binary coding "1" and "0" respectively as "not passable" and "allowable to pass".

[0054] Compared with the prior art, the present application has the following advantages:

[0055] 1. The application is directed to the mobile robot MTIPP (multi-target point inspection path planning) problem in the indoor kernel environment, which comprehensively considers the path length, cumulative radiation dose and turning energy consumption, and establishes a double-layer mathematical model and a double-layer hybrid algorithm. The inspection path obtained by the planning has better safety and lower global cost compared with the path planning in the prior art which only considers a single factor.

[0056] 2. In the double-layer mathematical model, the upper model (inspection path advantage and disadvantage evaluation model) is used to quantitatively evaluate the advantages and disadvantages of the inspection target point traversal sequence, and the lower model (pairwise path cost calculation model) proposes a calculation method for the three-dimensional cost of the pairwise path. In the double-layer hybrid algorithm, the upper algorithm (ACO-GA hybrid algorithm) is used to determine the optimal inspection target point traversal sequence (inspection path), and the lower algorithm (improved A*) is used to calculate the three-dimensional cost matrix (cost map) of the pairwise path. The upper algorithm and the lower algorithm are iteratively run (cyclically run), and the cost map calculated by the lower algorithm is returned to the upper algorithm to guide the upper algorithm to find a lower-cost inspection path, until the upper algorithm converges, i.e., the globally optimal inspection path is obtained.

[0057] 3. The improved A* algorithm introduces multiple constraint conditions in the heuristic function, and compared with the traditional A* algorithm, the total cost of the pairwise path is lower, which makes it more suitable for solving the mobile robot MTIPP problem in the indoor kernel environment.

[0058] 4. The ACO-GA hybrid algorithm uses the coarse optimal solution of the inspection path obtained by the GA algorithm to process the initial pheromone non-uniformity in the ACO algorithm, which improves the convergence of the ACO algorithm in the inspection target point traversal sequence optimization process. Moreover, the heuristic function, selection probability function and pheromone update rule of the ACO algorithm are designed by comprehensively considering the path length, cumulative radiation dose and turning energy consumption, which has dual advantages in solving quality and convergence speed, and is especially suitable for solving the mobile robot MTIPP problem in the indoor kernel environment.

[0059] The application will be further described in combination with the drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 The technical roadmap of the application is shown in the figure;

[0061] Figure 2 The mobile robot MTIPP problem in the indoor kernel environment is shown in the figure;

[0062] Figure 3 The eight-direction search strategy adopted by the mobile robot is shown in the figure;

[0063] Figure 4This is a comparison image of the initial pheromone non-uniformity treatment in the ACO algorithm before and after the treatment.

[0064] Figure 5 Box plots showing the comparison of the four algorithms in scenario 1 under the shortest path condition;

[0065] Figure 6 Box plot comparing the four algorithms in scenario 1 under the minimum radiation dose path;

[0066] Figure 7 Box-line comparison diagram of the four algorithms in scenario 1 under the minimum turning energy consumption path;

[0067] Figure 8 Box plots comparing the paths formed by the four algorithms in scenario 1 under three constraints;

[0068] Figure 9 Box plots showing the comparison of the four algorithms in scenario 2 under the shortest path condition;

[0069] Figure 10 Box plots showing the comparison of four algorithms in scenario 2 under the minimum radiation dose path;

[0070] Figure 11 Box-line comparison diagram of the four algorithms in scenario 2 under the minimum turning energy consumption path;

[0071] Figure 12 Box plots comparing the paths formed by the four algorithms in scenario 2 under the consideration of three constraints;

[0072] Figure 13 The inspection path map is obtained by using the ACO-GA-A*, GA-ACO-A*, and GA-A* algorithms in scenario 1.

[0073] Figure 14 This is the optimal inspection path diagram obtained using ACO-A* in scenario 1;

[0074] Figure 15 The inspection path map obtained using the ACO-GA-A* algorithm in scenario 2;

[0075] Figure 16 The inspection path map is obtained by using the GA-ACO-A* and GA-A* algorithms in scenario 2.

[0076] Figure 17 The inspection path map is obtained using the ACO-A* algorithm in scenario 2;

[0077] Figure 18 A comparison of the convergence curves of the ACO-GA-A*, GA-ACO-A*, GA-A*, and ACO-A* algorithms when they obtain the optimal solution in scenario 1;

[0078] Figure 19 Figure 4 is a comparison chart of convergence curves of ACO-GA-A*, GA-ACO-A*, GA-A* and ACO-A* algorithms in obtaining optimal solutions under scenario 2. DETAILED DESCRIPTION

[0079] Example 1

[0080] The method for multi-target point inspection path planning of a mobile robot in an indoor nuclear environment is provided, wherein a map of the indoor nuclear environment is prior information. The map is in the form of a grid map, and the grid in the grid map is a square. Each grid in the grid map contains radiation field information and obstacle information. The radiation field information is the average radiation dose rate recorded in each grid, and the obstacle information is identified by binary coding "1" and "0" in each grid to represent "non-passable" and "allowable to pass", respectively.

[0081] In combination Figures 2-3 The mobile robot MTIPP problem in an indoor nuclear environment is described. Figure 2 The grid map of the indoor nuclear environment is shown in FIG. 1. In order to more intuitively display the radiation field information, different colors are used to distinguish different numerical intervals of the average radiation dose rate (the average radiation dose rate gradually increases according to the change rule of light blue-beige-bright yellow-pink-deep red). In order to more intuitively display the obstacle information, black is used to represent the grid "non-passable", and other grids except black are "allowable to pass". Figure 2 In FIG. 1, 11 inspection target points (T1, T2, T3…T 11 The 11 inspection target points are connected in a specific traversal order to form a Hamiltonian circuit. Figure 3 In FIG. 1, the mobile robot adopts an eight-direction search strategy for path search, i.e., when the mobile robot moves from a current grid point to a neighboring grid, it can only move along the straight line path from the center point of the current grid to the center point of the neighboring grid.

[0082] As shown in FIG. 2, the method is as follows: Figure 1

[0083] S01, mathematical modeling of multi-target point inspection path:

[0084] A, establishing an inspection path evaluation model: the multi-target point inspection task needs to establish a Hamiltonian circuit that passes through each inspection target point only once (the start and end inspection target points are the same point, which is equivalent to passing through twice, so it is the only exception), and finally returns to the starting point. Considering the path length, cumulative radiation dose and turning energy consumption in three dimensions, an inspection path evaluation model (this model aims to minimize) or an inspection target point traversal order evaluation model is established, see formula 1; ​

[0085] Formula 1:

[0086] wherein λ is the total value of three dimensions of all pairs of paths in the inspection path (the path between two inspection target points connected in front and back is called a pair of paths); T is a set of inspection target points, T = {T1, T2, T3, …, Tn}, and the subscript n is the total number of inspection target points; is a path length cost matrix between inspection target points T i and T j , is a cumulative radiation dose cost matrix between inspection target points T i and T j , is a turning energy cost matrix between inspection target points T i and T j ; τi ,j is a decision variable, if the value is 1, it indicates that the path between inspection target point T i and inspection target point T j is selected, if the value is 0, it indicates that the path between inspection target point T i and inspection target point T j is not selected; ω1 is a weight coefficient of the importance of path length, ω2 is a weight coefficient of the importance of cumulative radiation dose, and ω3 is a weight coefficient of the importance of turning energy.

[0087] B, a pair of path cost calculation model is established: T i and T j are any two inspection target points, it is assumed that the grid set successively passed by the path with the minimum total value of three dimensions from T i to T j is P, P = {p1, p, p3…p m}, then the grid where T i is located is p1, and the grid where T j is located is p m ; considering the three dimensions of path length, cumulative radiation dose and turning energy consumption, a pair of path cost calculation model (the model aims to find the minimum value) is established, as shown in Formula 2.

[0088] Formula 2:

[0089] wherein M(P) is the total value of three dimensions of a pair of paths (the path between two inspection target points is called a pair of paths) T i to T j , L(P) is the path length cost value of T i to T j , R(P) is the cumulative radiation dose cost value of T i to Tj The cumulative radiation dose cost, E(P) is T i To T j The energy consumption cost of turning; ε1 is the weighting coefficient of the steering angle variable, ε2 is the weighting coefficient of the number of turns, and l(pa-1,pa) is the grid p α-1 With grid p α The line connecting the center points of pa, pa +1 ) is the raster p α With grid p α+1 The line connecting the center points, ψ[l(pa -1 ,pa),l(pa,pa +1 )] is line segment l(pα) -1 ,pa) and line segment l(pa,pa) +1 The steering angle variable is Nturn, where Nturn is the number of turns; (xα, yα) is the grid p. α (Map coordinate system) coordinates, (xα) +1 ,yα +1 ) is the raster p α+1 The coordinates (in the map coordinate system), (xα) -1 ,yα -1 ) is the raster p α-1 The coordinates (in the map coordinate system); R(pα,pα) +1 For mobile robots from grid p a Drive to the grid p a+1 The cumulative radiation dose received The average radiation dose rate of the grid pα, For grid pα +1 The average radiation dose rate, l is the grid pα to pα +1 The distance between grids pα and pα +1 If the edges are adjacent, then l equals 1 times the grid edge length. If grids pα and pα +1 If the endpoints are in contact, then it equals Double the grid side length, v is the distance from grid pα to grid pα. +1 Velocity (initial velocity of the mobile robot at the position of grid pα and grid pα) +1 The final velocity at each location is consistent, and the speed of the mobile robot remains constant during movement. The average radiation dose rate of a grid refers to the average radiation dose rate within the three-dimensional space formed by the grid at a height of 1m above the ground.

[0090] S02, Obtain the inspection path based on the upper-level algorithm:

[0091] A. Using the GA algorithm to obtain a coarse-optimized solution for the inspection path:

[0092] The chromosome in the GA algorithm corresponds to the inspection target point traversal sequence (i.e., a candidate solution), the gene in the GA algorithm corresponds to the inspection target point (i.e., an optimization variable), and the fitness in the GA algorithm corresponds to the inspection target point traversal sequence optimization evaluation function (see Formula 1). The GA algorithm process is as follows:

[0093] a1, real number coding: according to the numbers 1, 2, 3…n, the inspection target points are deployed in turn, indicating that the mobile robot should traverse n inspection target points, and a random inspection target point traversal sequence is generated as the initial population;

[0094] a2, create a taboo table: for each real gene coding, the selected taboo table is marked as 1, and the unselected taboo table is marked as 0;

[0095] a3, population initialization: set the evolution generation counter G and the maximum evolution generation G max , randomly generate an initial population;

[0096] a4, individual evaluation: evaluate all inspection target point traversal sequences according to the Hamiltonian circuit evaluation function (see Formula 1);

[0097] a5, individual selection: select multiple inspection target point traversal sequences with the smallest generation value as operators, which are applied to the existing population, so that individuals with good genes are retained for the next generation;

[0098] a6, crossover operation: according to the preset crossover probability P c , select individuals from the existing population, generate new offspring individuals through the crossover operator, and thus generate new inspection target point traversal sequences;

[0099] a7, evolution termination judgment: when G≤G max , repeat steps a3-a6; when G>G max , the evolution is terminated, and the individual with the smallest generation value in the population is the coarse optimization solution of the inspection path, denoted as path GA .

[0100] B, initial pheromone non-uniformization processing in ACO algorithm: the coarse optimization solution of the inspection path obtained by the GA algorithm is used for initial pheromone non-uniformization processing in the ACO algorithm, the processing method is shown in Formula 3, and the processing effect is shown in Figure 4 ; Formula 3 aims to improve the convergence of the ACO algorithm in the inspection target point traversal sequence optimization process through the guidance of non-uniform pheromones;

[0101] Formula 3:

[0102] In the formula, τ 0 (i,j) represents the inspection target point T iwith T j the initial pheromone concentration after the non-uniform treatment between T i with T j is a pair of paths, C is the initial pheromone weight value, τ is the set initial pheromone concentration (recommended to be set to 0.5), s.t. is the constraint condition; formula 3 represents that if the inspection target points T i and T j belong to the inspection target points in path GA , the initial pheromone weight value is added to the initial pheromone concentration as the initial pheromone concentration after the non-uniform treatment, and if the inspection target points T i and T j do not belong to the inspection target points in path GA , the set initial pheromone concentration 0.5 is taken as the initial pheromone value after the non-uniform treatment.

[0103] C, optimize the heuristic function of ACO algorithm: consider the influence of path length, cumulative radiation dose and turning energy consumption on the heuristic function, and introduce an adaptive adjustment coefficient to balance the convergence speed and global search ability of ACO algorithm, construct the heuristic function of ACO algorithm, as shown in formula 4;

[0104] Formula 4:

[0105] In the formula, η t (i,j) is the heuristic function value from the inspection target point T i to T j in the tthiteration, the larger the value, the better the corresponding path, and the smaller the value, the worse the corresponding path; I c is the current iteration number, I max is the maximum iteration number; ξ is the adaptive adjustment coefficient; ω1 is the weight coefficient of path length value, ω2 is the weight coefficient of cumulative radiation dose value, ω3 is the weight coefficient of turning energy consumption value,

[0106] is the path length cost matrix from the inspection target point T i to T j , is the cumulative radiation dose cost matrix from the inspection target point T i to T j , is the turning energy consumption cost matrix from the inspection target point T i to T j ;

[0107] D, optimize the transition probability function of ACO algorithm: comprehensively consider the pheromone concentration and heuristic function, construct the transition probability function of ACO algorithm, as shown in formula 5;

[0108] Equation 5:

[0109] In the formula, P t (i,j) is the selection probability of the tth iteration from the inspection target point T i to T j ; α is the information importance factor, β is the heuristic function importance factor; τ t (i,j) represents the pheromone concentration of the tth iteration from the inspection target point T i to T j ; τ t (i,u) represents the pheromone concentration of the tth iteration from the inspection target point T i to T u ; T t ' is the set of untraversed inspection target points in the tth iteration; the denominator in the equation of Equation 5 is the sum of the selection probabilities of the tth iteration from the inspection target point T i to all untraversed inspection target points; if the inspection target point T j in Equation 5 is an untraversed inspection target point, the selection probability calculation of T i to T j is performed, and if the inspection target point T j has been traversed, the selection probability of T i to T j is 0.

[0110] E, the pheromone update rule of the optimized ACO algorithm: considering the path length and the cumulative radiation dose (suffered by the mobile robot) and the turning energy consumption, the pheromone update rule of the ACO algorithm is optimized, as shown in Equation 6; thus, the construction of the upper algorithm (ACO-GA hybrid algorithm) is completed, and the fine optimization solution of the inspection path is obtained by running the upper algorithm;

[0111] Equation 6:

[0112] In the formula, τt +1 (i,j) represents the pheromone concentration of the t+1th iteration from the inspection target point T i to T j (the path), τt(i,j) represents the pheromone concentration of the tth iteration from the inspection target point T i to T j (the path), ρ is the pheromone evaporation factor, Q is the pheromone deposition factor, υ t-mi is the total three-dimensional value of the optimal inspection path in the tth iteration (υ value is the smallest), υ t-ma is the total three-dimensional value of the worst inspection path in the tth iteration (υ value is the largest); I tThe optimal inspection path for the tth iteration; υ is the total dimension value of the inspection path; set the inspection target points in the inspection path I as n, then "Ti, Ti +1 " are two inspection target points in the inspection path I, and "Tn, T1" are two inspection target points in the inspection path I, the requirement of the inspection path I is that each inspection target point is passed through only once (the start point and the end point inspection target points are the same point, which is equivalent to passing through twice, so it is the only exception) and finally returns to the start point; is the path length cost matrix between the inspection target points T i and T i+1 , is the cumulative radiation dose cost matrix between the inspection target points T i and T i+1 , is the turning energy consumption cost matrix between the inspection target points T i and T i+1 , is the path length cost matrix between the inspection target points T n and T1, is the cumulative radiation dose cost matrix between the inspection target points T n and T1, is the turning energy consumption cost matrix between the inspection target points T n and T1.

[0113] S03, constructing the lower algorithm to calculate the paired path:

[0114] Considering the path length and the cumulative radiation dose and the turning energy consumption, the heuristic function of the A*(improved) algorithm is established, and all paired paths in the fine optimization solution of the inspection path are planned through the A*(improved) algorithm. The start point and the end point of the paired path are respectively named as the starting inspection target point and the end inspection target point. According to the three dimension values involved in the paired path evaluation model (formula 2), the heuristic function of the A*(improved) algorithm is shown in formula 7. Thus, the construction of the lower algorithm (A*(improved) algorithm) is completed, and the specific line and the three dimension cost matrix of all paired paths in the fine optimization solution of the inspection path are obtained by running the lower algorithm multiple times;

[0115] Formula 7:

[0116] In the formula, (g(p)+h(p)) is equivalent to L(P) in formula 2, which is used to represent the path length value between two inspection target points; R(p 1-m ) is equivalent to R(P) in formula 2, which is used to represent the cumulative radiation dose value between two inspection target points; E(p 1-m) is equivalent to E(P) in formula 2, and is used to represent the turning energy consumption value between two inspection target points; f(x) is the three-dimensional value between the starting inspection target point and the end inspection target point; g(p) is the path length value of the starting inspection target point to the current grid, h(p) is the path length value of the current grid to the end inspection target point, (x i ,y i ) is the coordinate of the current grid pi, (x m ,y m ) is the coordinate of the grid pm where the end inspection target point is located (the subscript m represents that there are m grids between the starting inspection target point and the end inspection target point).

[0117] S04, the two-layer algorithm is run to obtain a global optimal solution:

[0118] Each time the "inspection path fine optimal solution" obtained by running the upper layer algorithm is marked in order as A1, A2, A3…Az; each time the "three-dimensional cost matrix of all pairs of paths in the inspection path fine optimal solution" obtained by running the lower layer algorithm is marked in order as C1, C2, C3…Cz; the results A1 obtained by running the upper layer algorithm for the first time and the results C1 obtained by running the lower layer algorithm for the first time are substituted into formula 1 to calculate the three-dimensional total cost value corresponding to A1, which is denoted as λ1.

[0119] A1 is taken as the initial population of the GA algorithm, and the steps S02-S03 are repeatedly run to obtain A2 and C2; A2 and C2 are substituted into formula 1 to calculate the three-dimensional total cost value corresponding to A2, which is denoted as λ2; thus the first iteration is completed; λ1 and λ2 are compared, if λ2≥λ1, then A1 corresponding to λ1 is the global optimal solution of the inspection path, if λ2<λ1, then A2 is put into the next iteration.

[0120] A2 is taken as the initial population of the GA algorithm, and the steps S02-S03 are repeatedly run to obtain A3 and C3; A3 and C3 are substituted into formula 1 to calculate the three-dimensional total cost value corresponding to A3, which is denoted as λ3; thus the second iteration is completed; λ2 and λ3 are compared, if λ3≥λ2, then A2 corresponding to λ2 is the global optimal solution of the inspection path, if λ3<λ2, then A3 is put into the next iteration.

[0121] In this way, the global optimal solution of the inspection path is finally obtained.

[0122] As a preferred solution, in the S03 step, the A*(improved) algorithm further comprises a high radiation dose rate danger zone avoidance strategy; when the mobile robot passes through a grid area whose radiation dose rate exceeds the bearing limit of the mobile robot, the mobile robot will be damaged by radiation and even cause component failure, thus it is necessary to avoid such grid area; in the path search process, when an ant senses that the radiation dose rate of the next grid (of the current grid) exceeds the bearing limit of the mobile robot, the ant defines the radiation dose rate surrogate value of the next grid as infinite, so as to prompt the ant to find other passable grids, and refer to formula 8.

[0123] Formula 8:

[0124] Wherein, R(pi) is the radiation dose rate surrogate value of the grid pi, Inf is infinite, and RImax is the maximum radiation dose rate that can be borne by the mobile robot; formula 8 indicates that if the radiation dose rate of the grid pi does not exceed the maximum radiation dose rate that can be borne by the mobile robot, the radiation dose rate of the grid pi is not changed, and if the radiation dose rate of the grid pi exceeds the maximum radiation dose rate that can be borne by the mobile robot, the radiation dose rate of the grid pi is defined as infinite.

[0125] Technical effect analysis of the application: the ACO-GA-A*(double-layer mixed algorithm, i.e., the algorithm composed of the upper algorithm and the lower algorithm) is compared with other algorithms in two indoor core environments with different complexities, and the performance of the ACO-GA-A* in sensitivity, reliability and convergence is verified.

[0126] I. Sensitivity analysis:

[0127] When solving the MTIPP problem, three constraint conditions of path length, risk degree and turning energy consumption are comprehensively considered. Therefore, it is necessary to verify whether the optimal path preferred by the user can be found by adjusting the cost coefficient weight (ω1, ω2, ω3). For this purpose, according to the actual working condition requirements of the core environment mobile robot, the sensitivity of the ACO-GA-A* double-layer mixed algorithm and other algorithms under four groups of weights is compared, wherein the coefficient weight combination (1, 0, 0) represents the shortest path; the coefficient weight combination (0, 1, 0) represents the minimum cumulative radiation dose path; the coefficient weight combination (0, 0, 1) represents the minimum turning energy consumption path; and the coefficient weight combination (1, 1, 1) represents the optimal path considering the three constraints at the same time. The comparative algorithms under different coefficient weight combinations are run 20 times under two indoor core environment scenes (scene 1 and scene 2) with different complexities. The box line comparison chart of different preferred paths under different algorithms in scene 1 is shown in FIG. 1. Figures 5-8 The box line comparison chart of different preferred paths under different algorithms in scene 2 is shown in FIG. 2. Figures 9-12

[0128] Analysis​Figures 5-12 It can be seen that: 1. The ability of ACO-GA-A* to find the shortest path in different nuclear environment scenarios is better than that of other algorithms, and the solution set is relatively concentrated (small deviation). 2. The ability of ACO-GA-A* to find the minimum cumulative radiation path in different nuclear environment scenarios is better than that of other algorithms, and the solution set is relatively concentrated (small deviation). 3. When considering three constraint conditions, the total cost of the optimal inspection path planned by ACO-GA-A* in different scenarios is less than that of other hybrid algorithms, indicating that it has obvious advantages in solution quality. The above analysis shows that the changes in weight coefficients and environmental complexity will not affect the effectiveness of ACO-GA-A*.

[0129] II. Reliability analysis

[0130] The reliability of ACO-GA-A* was analyzed based on simulation of two different complexity indoor nuclear environment scenarios (Scenario 1 and Scenario 2). The single-objective standardization method was used to determine the weight combination of the objective function. The reliability data of different algorithms under Scenario 1 and Scenario 2 were obtained by simulating 15 times under Scenario 1 and Scenario 2, respectively, as shown in Table 1. The optimal inspection path planned by different algorithms under Scenario 1 is shown in FIG. 1 (since the optimal inspection paths planned by ACO-GA-A*, GA-ACO-A* and GA-A* under Scenario 1 are consistent, only the optimal inspection path planned by ACO-GA-A* is shown in FIG. 1). Figures 13-14 Figure 13 The optimal inspection path planned by ACO-A* under Scenario 1 is shown in FIG. 2), and the optimal inspection path planned by different algorithms under Scenario 2 is shown in FIG. 3 (since the optimal inspection paths planned by GA-ACO-A* and GA-A* under Scenario 2 are consistent, only the optimal inspection path planned by GA-ACO-A* is shown in FIG. 3). Figure 14 Figures 15-17 The optimal inspection path planned by ACO-A* under Scenario 2 is shown in FIG. 4), and the optimal inspection path planned by ACO-GA-A* and ACO-A* under Scenario 2 is shown in FIG. 5 and FIG. 6, respectively. Figure 16 Figure 15 Figure 17

[0131] Table 1: Reliability data of different algorithms under different scenarios

[0132]

[0133] ​​​​​As can be seen from Table 1, in different complex nuclear environment scenarios, the average total cost and its standard deviation of the path planned by ACO-GA-A* are significantly better than those of other algorithms, and the more complex the environment, the more obvious the advantage. The probability of obtaining the optimal path under scenario 1 and scenario 2 reaches 100% and 30% respectively, which is significantly higher than that of other algorithms. Therefore, ACO-GA-A* algorithm has higher reliability and robustness in solving MTIPP problem than other algorithms. In terms of calculation time, ACO-GA-A* and GA-ACO-A* take about the same time, and slightly longer than ACO-A* and GA-A*. The reason is that it takes more time cost to optimize the initial pheromone distribution of ACO by GA or to optimize the initial population of GA by ACO. However, considering that the MTIPP problem is offline path planning before the mobile robot enters the inspection site, it is acceptable to obtain a better path with a small amount of time cost.

[0134] In Figures 13-17 , in order to more intuitively display the radiation field information, different colors are used to distinguish different numerical intervals of the average radiation dose rate (the average radiation dose rate gradually increases according to the blue-yellow-red change rule), and in order to more intuitively display the obstacle information, black is used to represent the grid "not passable", and other grids except black are "allowed to pass". From Figures 13-17 it can be seen that the optimal path planned by ACO-GA-A* under different nuclear environment scenarios has good traversal sequence, no path intersection, and the formed path does not pass through the high radiation danger area (red area).

[0135] Ⅲ. Convergence analysis

[0136] The convergence curve comparison chart of different algorithms when obtaining the optimal solution under scenario 1 is shown in Figure 18 , and the upper right corner of Figure 18 is a local magnification in Figure 18 .The convergence curve comparison chart of different algorithms when obtaining the optimal solution under scenario 2 is shown in Figure 19 , and the upper right corner of Figure 19 shows the algorithm represented by the line of different colors (also the same as Figure 18 ).

[0137] From Figures 18-19As can be seen, the ACO-GA-A* algorithm converges to the optimal solution faster, while the other algorithms converge relatively slowly and cannot plan the minimum total cost inspection path. With the increase of the complexity of the environment, the performance of the ACO-A* algorithm decreases rapidly, especially in terms of the optimal solution. In addition, the number of iterations of the GA-A* algorithm increases most significantly with the increase of the complexity of the environment. By comparing the convergence curves of ACO-GA-A* and ACO-A*, it can be confirmed that the enhancement of GA to the initial pheromone guidance of ACO has a significant effect on improving the convergence speed. By comparing the convergence of GA-ACO-A* and GA-A*, it can be seen that the initial population optimization of GA based on ACO significantly improves the convergence speed of GA. In summary, the ACO-GA-A* algorithm is more likely to find the optimal solution.

Claims

1. A method for multi-target point inspection path planning of a mobile robot in a kernel environment, characterized in that, The grid map of the indoor kernel environment is prior information, and each grid in the grid map contains radiation field information and obstacle information; The method is as follows: S01, multi-target point inspection path mathematical modeling: considering path length, cumulative radiation dose and turning energy consumption, an inspection path evaluation model and a pair of path evaluation model are respectively established; S02, obtaining the inspection path based on the upper algorithm: the GA algorithm is used to obtain the coarse optimization solution of the inspection path; The initial pheromone in the ACO algorithm is processed by using the coarse optimization solution of the inspection path; the heuristic function and the selection probability function of the ACO algorithm are constructed by considering the path length, the cumulative radiation dose and the turning energy consumption, and the pheromone update rule of the ACO algorithm is optimized; the upper algorithm is constructed and run to obtain the fine optimization solution of the inspection path; S03, constructing the lower algorithm to calculate the pair of paths: the heuristic function of the A* algorithm is established by considering the path length, the cumulative radiation dose and the turning energy consumption; the lower algorithm is constructed and run to plan all the pairs of paths in the fine optimization solution of the inspection path; S04, the optimal solution obtained by the two-layer algorithm: the fine optimization solutions of the inspection path obtained by the current iteration and the last iteration are compared by using the inspection path evaluation model, if the fine optimization solution of the inspection path obtained by the current iteration is better, the fine optimization solution of the inspection path obtained by the current iteration is put into the next iteration; If the fine optimization solution of the inspection path obtained by the current iteration is worse or consistent, the iteration is ended, and the fine optimization solution of the inspection path obtained by the last iteration is output.

2. The method of claim 1, wherein the method further comprises: determining a first path from the first target point to the second target point; determining a second path from the second target point to the third target point; and determining a third path from the third target point to the first target point. The step S01 is described in detail as follows: A, establishing the inspection path evaluation model: a Hamilton circuit that each inspection target point passes through only once and finally returns to the starting point is needed for the multi-target point inspection task, and the inspection path evaluation model, or the inspection target point traversal order evaluation model, is established by considering three dimensions of path length, cumulative radiation dose and turning energy consumption, refer to formula 1; Formula 1: wherein λ is the total value of all pairs of paths in the inspection path; T is a set of inspection target points, T = {T1, T2, T3, …, Tn}, and subscript n is the total number of inspection target points; n}; is a path length cost matrix between inspection target points T i and T j , is a cumulative radiation dose cost matrix between inspection target points T i and T j , is a turning energy consumption cost matrix between inspection target points T i and T j ; τ i,j is a decision variable, if the value is 1, it indicates that the path between inspection target point T i and inspection target point T j is selected, if the value is 0, it indicates that the path between inspection target point T i and inspection target point T j is not selected; ω1 is a weight coefficient of the path length, ω2 is a weight coefficient of the cumulative radiation dose, and ω3 is a weight coefficient of the turning energy consumption. B, Establish a pairwise path cost calculation model: T i and T j Let T be any two inspection target points. i To T j Let P be the set of grid cells that the path with the minimum total value in the three dimensions passes through in sequence, where P = {p1, p2, p3, ..., p4}. m }, then T i If the grid cell is p1, then T j The grid cell it belongs to is p m Considering three dimensions—path length, cumulative radiation dose, and turning energy consumption—a pairwise path cost calculation model is established, as shown in Formula 2. Equation 2: Wherein, M(P) is the pair path T i to T j three-dimensional total value, L(P) is the path length value of T i to T j , R(P) is the cumulative radiation dose value of T i to T j , E(P) is the turning energy consumption value of T i to T j ; ε1 is the weight coefficient of the steering angle variable, ε2 is the weight coefficient of the steering frequency, l(p a -1,p a ) is the center line of the grid p α-1 and the grid p α , l(p a ,p a+1 ) is the center line of the grid p α and the grid p α+1 , ψ[l(p a-1 ,p a ), l(p a ,p a+1 )] is the steering angle variable of the line segment l(p a-1 ,p a ) and the line segment l(p a ,p a+1 ), N turn is the steering frequency; (x α ,y α ) is the coordinate of the grid p α , (x α+1 ,y α+1 ) is the coordinate of the grid p α+1 , (x α-1 ,y α-1 ) is the coordinate of the grid p α-1 ; R(p α ,p α+1 ) is the cumulative radiation dose received by the mobile robot from the grid p a to the grid p a+1 , is the average radiation dose rate of the grid p α , is the average radiation dose rate of the grid p α+1 , l is the distance from the grid p α to p α+1 , if the grid p α is adjacent to p α+1 edge, then l is equal to 1 times the grid edge length, if the grid p α is in contact with p α+1 end point, then l is equal to times the grid edge length, v is the speed from the grid p α Move to grid p α+1 Speed of the velocity.

3. The method of claim 2, wherein the method further comprises: determining a path of the mobile robot to each of the plurality of target points; and determining a path of the mobile robot to each of the plurality of target points based on the determined path of the mobile robot to each of the plurality of target points. The step S02 is described in detail as follows: A, obtaining the coarse optimization solution of the inspection path by using the GA algorithm: The chromosome in the GA algorithm corresponds to the inspection target point traversal order, the gene in the GA algorithm corresponds to the inspection target point, and the fitness in the GA algorithm corresponds to the inspection target point traversal order evaluation function; The GA algorithm process is as follows: a1, real number coding: the inspection target points are deployed in the order of 1, 2, 3…n, indicating that the mobile robot should traverse n inspection target points, and a random inspection target point traversal order is generated as the initial population; a2, creating a taboo table: for each real gene coding, the selected taboo table is marked as 1, and the unselected taboo table is marked as 0; a3, population initialization: set the evolutionary generation counter G and the maximum evolutionary generations G max , randomly generate the initial population; a4, individual evaluation: all inspection target point traversal orders are evaluated according to the Hamilton circuit evaluation function; a5, individual selection: select multiple inspection target point traversal orders with the smallest generation value as operators, and apply them to the existing population, so that individuals with good genes are retained for the next generation; a6, crossover operation: according to a preset crossover probability P c Selecting individuals from the existing population, generating new offspring individuals through the crossover operator, and thus generating a new inspection target point traversal sequence; a7, evolution termination judgment: when G≤G max , repeat a3-a6 sub-steps; when G>G max , evolution termination, the individual with the minimum fitness value in the population is the coarse optimization solution of the inspection path, denoted as path GA ; B, initial pheromone non-uniformization processing in the ACO algorithm: the initial pheromone in the ACO algorithm is processed by using the coarse optimization solution of the inspection path obtained by the GA algorithm, as shown in formula 3; Equation 3: where τ 0 (i,j) represents the initial pheromone concentration after the non-uniformization process between the inspection target points T i and T j , the inspection target points T i and T j are paired paths, C is the initial pheromone weighting value, τ is the set initial pheromone concentration, and s.t. is the constraint condition; C, optimizing the heuristic function of ACO algorithm: considering the influence of path length, cumulative radiation dose and turning energy consumption on the heuristic function, and introducing adaptive adjustment coefficient to balance the convergence speed and global search ability of ACO algorithm, the heuristic function of ACO algorithm is constructed, as shown in formula 4; Equation 4: wherein η t (i,j) is the heuristic function value from the inspection target point T i to T j in the tth iteration; I c is the current iteration number, I max is the maximum iteration number; ξ is an adaptive adjustment coefficient; ω1 is a weight coefficient of the path length cost value, ω2 is a weight coefficient of the cumulative radiation dose cost value, ω3 is a weight coefficient of the turning energy consumption cost value, is a path length cost matrix from the inspection target point T i to T j , is a cumulative radiation dose cost matrix from the inspection target point T i to T j , is a turning energy consumption cost matrix from the inspection target point T i to T j . D, optimizing the transition probability function of ACO algorithm: considering the influence of pheromone concentration and heuristic function, the transition probability function of ACO algorithm is constructed, as shown in formula 5; Equation 5: where P t (i,j) is the selection probability of the tth iteration from the inspection target point T i to T j ; a is an information element importance factor, and β is a heuristic function importance factor; τ t (i,j) represents the information element concentration of the tth iteration from the inspection target point T i to T j ; τ t (i,u) represents the information element concentration of the tth iteration from the inspection target point T i to T u ; and T t ' is the set of untraversed inspection target points in the tth iteration. E, optimizing the pheromone update rule of ACO algorithm: considering the influence of path length, cumulative radiation dose and turning energy consumption, the pheromone update rule of ACO algorithm is optimized, as shown in formula 6; thus the construction of upper algorithm is completed, and the fine optimization solution of inspection path is obtained by running the upper algorithm; Equation 6: where τ t+1 (i,j) represents the pheromone concentration from the inspection target point T i to T j in the t+1th iteration, τ t (i,j) represents the pheromone concentration from the inspection target point T i to T j in the tth iteration, p is the pheromone evaporation factor, Q is the pheromone deposition factor, υ t-mi is the total fitness value of the optimal inspection path in the tth iteration, υ t-ma is the total fitness value of the worst inspection path in the tth iteration; and I t is the optimal inspection path in the tth iteration. υ is the total dimension value of the inspection path; assuming that there are n inspection target points in the inspection path, then "T i i+1 ” are two inspection target points in the inspection path, then "T n i T1” are two inspection target points in the inspection path, and the inspection path requires that each inspection target point is passed only once and finally returns to the starting point; is the path length cost matrix between the inspection target points T i and T i+1 , is the cumulative radiation dose cost matrix between the inspection target points T i and T i+1 , is the turning energy consumption cost matrix between the inspection target points T i and T i+1 , is the path length cost matrix between the inspection target points T n and T1, is the cumulative radiation dose cost matrix between the inspection target points T n and T1, is the turning energy consumption cost matrix between the inspection target points T n and T1.

4. The method of claim 3, wherein the method further comprises: determining a path of the mobile robot to each of the plurality of target points; and determining a path of the mobile robot to each of the plurality of target points based on the determined path of the mobile robot to each of the plurality of target points. S03 is described in detail as follows: Considering the influence of path length, cumulative radiation dose and turning energy consumption, the heuristic function of A* algorithm is established, and all pairs of paths in the fine optimization solution of inspection path are planned by A* algorithm. The starting point and the end point of the pair of paths are named as the starting inspection target point and the end inspection target point, respectively. According to the three dimension cost values involved in the pair of path evaluation model, the heuristic function of A* algorithm is shown in formula 7; thus the construction of lower algorithm is completed, and the specific route of all pairs of paths in the fine optimization solution of inspection path and the three dimension cost matrix are obtained by running the lower algorithm for multiple times; Equation 7: In the formula, (g(p)+h(p)) is equivalent to L(P) in formula 2, and is used to represent the path length value between two inspection target points; R(p 1-m ) is equivalent to R(P) in formula 2, and is used to represent the cumulative radiation dose value between two inspection target points; E(p 1-m ) is equivalent to E(P) in formula 2, and is used to represent the turning energy consumption value between two inspection target points; f(x) is the three-dimensional value between the starting inspection target point and the end inspection target point; g(p) is the path length value from the starting inspection target point to the current grid, h(p) is the path length value from the current grid to the end inspection target point, (x1,y1) is the coordinate of the grid p1 where the starting inspection target point is located, (x i ,y i ) is the coordinate of the current grid p i , and (x m ,y m ) is the coordinate of the grid p m where the end inspection target point is located.

5. The method of claim 4, wherein the method further comprises: determining a path for the mobile robot to travel to each of the plurality of target points in the indoor kernel environment; and determining a path for the mobile robot to travel between each of the plurality of target points in the indoor kernel environment. The A* algorithm in S03 also includes high radiation dose rate dangerous area avoidance strategy. During the path search process, when the ant perceives that the radiation dose rate of the next grid exceeds the bearing limit of the mobile robot, the radiation dose rate cost value of the next grid is defined as infinity, which prompts the ant to find other passable grids. Refer to formula 8; Equation 8: where R(p i ) is the radiation dose rate value of the grid p i , Inf is infinity, and R max I is the maximum radiation dose rate that the mobile robot can withstand.

6. The method of claim 5, wherein the method further comprises: determining a path of the mobile robot to each of the plurality of target points; and determining a path of the mobile robot to each of the plurality of target points based on the determined path of the mobile robot to each of the plurality of target points. S04 is described in detail as follows: Mark the "fine optimization solution of inspection path" obtained by running the upper algorithm each time as A1, A2, A3…Az in order; mark the "three dimension cost matrix of all pairs of paths in the fine optimization solution of inspection path" obtained by running the lower algorithm each time as C1, C2, C3…Cz in order; put the results obtained by running the upper algorithm for the first time A1 and the results obtained by running the lower algorithm for the first time C1 into formula 1 to calculate the three dimension total cost value corresponding to A1, which is recorded as λ1; Take A1 as the initial population of GA algorithm, repeat the steps of S02-S03 to obtain A2 and C2; put A2 and C2 into formula 1 to calculate the three dimension total cost value corresponding to A2, which is recorded as λ2; thus the first iteration is completed; compare λ1 with λ2, if λ2≥λ1, then A1 corresponding to λ1 is the global optimal solution of inspection path, if λ2<λ1, then put A2 into the next iteration; Take A2 as the initial population of GA algorithm, repeat the steps of S02-S03 to obtain A3 and C3; put A3 and C3 into formula 1 to calculate the three dimension total cost value corresponding to A3, which is recorded as λ3; thus the second iteration is completed; compare λ2 with λ3, if λ3≥λ2, then A2 corresponding to λ2 is the global optimal solution of inspection path, if λ3<λ2, then put A3 into the next iteration; Repeat the above process, and finally the global optimal solution of inspection path is obtained.

7. The method of claim 6, wherein the method further comprises: determining a path of the mobile robot to each of the plurality of target points; and determining a path of the mobile robot to each of the plurality of target points based on the determined path of the mobile robot to each of the plurality of target points. The grid in the grid map is a square; the radiation field information contained in each grid in the grid map is the average radiation dose rate recorded in each grid, and the obstacle information contained in each grid in the grid map is identified as "not passable" and "allowable to pass" respectively by binary coding "1" and "0" in each grid.

Images