Reinforcing bar binding robot path planning method and device based on improved ant colony algorithm

Abstract

The application discloses a reinforced bar binding robot path planning method and device based on an improved ant colony algorithm, and relates to the technical field of robot path planning.The technical scheme of the application firstly converts the reinforced bar net environment of robot operation into an electronic map containing a plurality of task points by considering the robot operation range and obstacles, secondly optimizes the heuristic function for two running modes of the guide rail type robot on the reinforced bar net, i.e., running along the guide rail reinforced bar and jacking up and translating perpendicularly to the guide rail reinforced bar, improves the transfer probability difference of the two running modes, reduces the jacking up and translating frequency, and solves the deadlock problem occurring during traversal by changing the target point selection range, so that the most efficient operation path traversing all the task points can be obtained.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of robot path planning, in particular to a rebar binding robot path planning method and device based on an improved ant colony algorithm. BACKGROUND

[0002] With the continuous advancement of urbanization, the demand for buildings and bridges is also increasing, and the traditional construction method cannot meet the requirements of construction productivity, quality, safety and sustainable development. Intelligent construction robots combining automation technology and robot technology are gradually introduced into various construction scenes as ideal labor forces, and a rebar binding robot has a wide application prospect in improving the production efficiency, binding standardization and construction safety of the rebar binding operation on a construction site.

[0003] When the construction robot is applied to a construction scene, it usually needs to complete three works, path recognition, path planning and behavior decision, wherein the path planning problem is a core problem of the autonomous operation of the construction robot, and the target is to find a collision-free optimal path for completing a specific task in a map containing static, dynamic obstacles or environmental interference. An application with an authorization announcement number CN116151036B discloses a rebar automatic binding path planning method. The method arranges the rebar intersection points in the scanning field of the binding robot in order, and judges whether to edit in reverse order by calculating the distance between adjacent points, so as to finally realize S-shaped path planning of all binding points. Although this method realizes the traversal path planning of the binding points in the working range of the rebar binding robot, the walking path of the binding robot on the whole rebar mesh is not planned. Therefore, the application provides a rebar binding robot path planning method and device based on an improved ant colony algorithm. SUMMARY

[0004] The application aims to provide a rebar binding robot path planning method and device based on an improved ant colony algorithm, which can fill the gap in the current construction robot path planning technology and method, and solve the obstacle avoidance and traversal path optimization problem of the binding robot on the rebar mesh.

[0005] To achieve the above-mentioned purpose, the application provides the following technical scheme: a rebar binding robot path planning method based on an improved ant colony algorithm, and the specific steps include:

[0006] S1. Simplify the working area into a plurality of task points according to the robot working range, establish an electronic map according to the grid method, and set the starting point, ending point and obstacle information of the path;

[0007] S2. Initialize the basic parameters of the ant colony algorithm, determine the initial pheromone concentration between adjacent task points, and reduce the pheromone concentration of adjacent points perpendicular to the direction of the rail reinforcement. The initial pheromone concentration is substituted into the pheromone update model;

[0008] S3. Generate an ant population, determine the number of ants in each iteration, and determine the initial task point coordinates;

[0009] S4. Place each ant at the starting point in the electronic map;

[0010] S5. Determine whether there are untraversed nodes around the current node of the ant. If there are, calculate the probability of the ant moving to the next node using the comprehensive heuristic function, and select the next target node using the roulette method. If there are no untraversed nodes, expand the range of selected nodes until an untraversed node appears;

[0011] S6. Determine whether the current ant has traversed all nodes. If yes, return to S4; if no, update the current node of the ant to the next target node determined in S5, return to S5, and continue the task node traversal;

[0012] S7. After all ants in this generation have completed the task node traversal, calculate the path length and the number of turns of each ant's traversal path according to the pheromone update model, obtain the optimal path of this iteration, i.e., the path with the shortest length, and record it. Then update the pheromone;

[0013] S8. Determine whether the number of path search iterations reaches the preset value. If yes, proceed to the next step; if no, return to S2;

[0014] S9. Calculate and compare the optimal traversal paths of each generation of ants, select the optimal path as the optimal path of this overall search, and save the relevant path information.

[0015] Further, the pheromone update model expression is:

[0016]

[0017]

[0018] C K (t)=uP k (t)+vH k (t);

[0019] In the formula, τ ij (t) represents the pheromone on the path from task point i to j at time t, ρ is the evaporation rate of pheromone, τ ij is the pheromone content before the path, The pheromone increment of the kth ant on the path (i, j) at time t, Q is the pheromone concentration, C K (t) is the comprehensive index of the pheromone of the kth ant on the path in the tth iteration, the smaller the index, the better the path, the more pheromone left, P k (t) is the length of the path, H k (t) is the number of times of moving along the direction perpendicular to the rail steel, u and v are adjustment parameters of two factors, which can take appropriate values according to needs.

[0020] Further, the expression of the comprehensive heuristic function in S5 is as follows:

[0021]

[0022] In the formula, is the comprehensive heuristic function, which represents the expectation of the kth ant moving from i node to j node; d i,j is the distance between i node and j node; λ is the distance coefficient, which is 1 when i node and j node are in the same rail steel direction, and is a value greater than 1 when i node and j node are not in the same rail steel direction, and the value is taken according to needs.

[0023] Further, the transition probability formula of the ant moving to the next node in S5 is:

[0024]

[0025] In the formula, C is the feasible region set of the next node of the ant; τ ij (t) is the pheromone value of the path (i, j); α is the pheromone heuristic factor, which determines the influence degree of pheromone concentration on path selection; β is the expectation heuristic factor, which determines the influence degree of heuristic information on path selection, τ iS (t) represents the pheromone value of the path (i, S) at time t, S belongs to a selectable node in C; η i,S (t) is the expectation of the path (i, S) at time t.

[0026] Further, the binding points on the reinforcement mesh in the working area are covered with the least number of circles according to the working range of the robot, and the center of each circle is taken as the task point of the robot work.

[0027] Further, the obstacle area formed by the irregularly shaped obstacle is converted into a regular pattern by cutting, and the working environment is converted into a grid map by combining the distribution of the task points and the obstacle area.

[0028] Further, the electronic map is established by using a simulation software, and the simulation software is MATLAB R2022b.

[0029] Further, the related parameters in the S1 are: the ant quantity m, the maximum iteration number N max , the information heuristic factor alpha, the expected heuristic factor beta, the pheromone volatilization factor rho, the pheromone concentration Q, and the starting point coordinates of the robot.

[0030] The application provides a steel bar binding robot path planning device based on an improved ant colony algorithm, and specifically comprises the following steps:

[0031] A simulation programming module is configured to simplify a working area into a plurality of task points according to a robot working range, establish an electronic map according to a grid method, and set starting point, end point and obstacle information of a path.

[0032] An initialization module is configured to initialize basic parameters of the ant colony algorithm, determine initial pheromone concentrations between adjacent task points, reduce the pheromone concentrations of adjacent points perpendicular to the steel bar direction of the guide rail, and substitute the initial pheromone concentrations into a pheromone update model.

[0033] An initial point determination module is configured to generate an ant population, determine the number of the ant population in each iteration, and determine initial working point coordinates.

[0034] A starting iteration module is configured to place each ant at a starting point in the electronic map.

[0035] A first judgment module is configured to judge whether there are untraversed nodes around the current node of the ant, and if so, calculate the probability of the ant moving to a next node by using a comprehensive heuristic function, and select the next target node by using a roulette method; if not, expand the range of the selected node until an untraversed node appears.

[0036] A second judgment module is configured to judge whether the current ant has traversed all nodes, and if so, return to S4; if not, update the parent node and return to S5 to continue the working node traversal.

[0037] An update module is configured to, after all ants in the generation complete the working node traversal, calculate the path length and the number of turns of the path traversed by each ant according to the pheromone update model, update the pheromone of the optimal path in the iteration, and obtain the optimal path of the iteration.

[0038] A third judgment module is configured to judge whether the iteration number of the path search reaches a preset value, and if so, execute the next step; if not, return to S2.

[0039] A calculation module is configured to calculate and compare the optimal traversal paths of the ants in each generation, select the optimal path as the optimal path of the overall search, and save the related path information.

[0040] With regard to one aspect of the present application, the present application provides a terminal device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, the improved ant colony algorithm-based rebar binding robot path planning method is adopted.

[0041] The present application has at least the following beneficial effects:

[0042] The present application firstly converts the rebar environment of the robot operation into an electronic map containing a plurality of task points by considering the robot operation range and obstacles, secondly optimizes the heuristic function for the two running modes of the guide rail robot on the rebar, i.e., running along the guide rail rebar and jacking and translating perpendicular to the guide rail rebar, improves the transfer probability difference of the two running modes, reduces the jacking and translating frequency, and solves the deadlock problem occurring during traversal by changing the target point selection range, so that an operation path with the highest efficiency for traversing all task points can be obtained, filling the gap in the current construction robot path planning technology and method, and solving the problem of obstacle avoidance and traversal path optimization of the binding robot on the rebar.

[0043] Of course, implementing any product of the present application does not necessarily require all the advantages described above to be achieved simultaneously. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 The flowchart of the path planning method of the present application;

[0045] Figure 2 The path planning result schematic diagram of the first embodiment of the present application;

[0046] Figure 3 The path planning result schematic diagram of the second embodiment of the present application. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0048] Please refer to Figure 1 The present application provides a technical solution: an improved ant colony algorithm-based rebar binding robot path planning method, and the specific steps include the following:

[0049] S1. According to the robot working range, the working area is simplified into several task points (i.e. working nodes), an electronic map is established in the simulation software according to the grid method, and the start point, end point and obstacle information of the path are set;

[0050] S2. The basic parameters of the ant colony algorithm are initialized, the initial pheromone concentration between adjacent task points is determined, the pheromone concentration of adjacent points perpendicular to the direction of the rail reinforcement is reduced, and the initial pheromone concentration is substituted into the pheromone update model;

[0051] Further, the pheromone update model expression is:

[0052]

[0053]

[0054] C K (t)=uP k (t)+vH k (t);

[0055] In the formula, τ ij (t) represents the pheromone on the path from task point i to j at time t, ρ is the evaporation rate of pheromone, τ ij is the pheromone content before the path, is the pheromone increment of the kth ant on the path (i,j) at time t, Q is the pheromone concentration, C K (t) is the comprehensive index of the pheromone of the kth ant on the path in the tth iteration, the smaller the index, the better the path, and the more pheromone left, P k (t) is the length of the path, H k (t) is the number of times of moving along the direction perpendicular to the rail reinforcement, u and v are adjustment parameters of the two factors, which can be appropriately valued according to the needs;

[0056] S3. Generate an ant colony, determine the number of ant colonies in each iteration, and determine the initial working point coordinates;

[0057] S4. Put each ant into the start point in the electronic map;

[0058] S5. Determine whether there are untraveled nodes around the current node of the ant, if there are, calculate the probability of the ant moving to the next node using the comprehensive heuristic function, and select the next target node using the roulette method; if there are not, expand the range of selected nodes until an untraveled node appears;

[0059] Further, the expression of the comprehensive heuristic function is as follows:

[0060]

[0061] In the formula, is the expected value of the kth ant moving from the i node to the j node; d i,j is the distance between the i node and the j node; λ is a distance coefficient, which is 1 when the i node and the j node are in the same direction of the guide rail, and is a value greater than 1 when the i node and the j node are not in the same direction of the guide rail, and is selected according to requirements;

[0062] Further, the transition probability formula of the ant moving to the next node is:

[0063]

[0064] In the formula, C is the feasible region set of the next node of the ant; τ ij (t) is the pheromone value of the path (i, j) at time t; α is a pheromone heuristic factor, which determines the influence degree of the pheromone concentration on path selection; β is an expected heuristic factor, which determines the influence degree of the heuristic information on path selection; τ iS (t) represents the pheromone value of the path (i, S) at time t, and S is a selectable node in C; η i,S (t) is the expected value of the path (i, S) at time t

[0065] S6. Determine whether the current ant has traversed all nodes, if yes, return to S4; if not, update the node where the ant is currently located to the next target node determined in S5, return to S5, and continue to traverse the work nodes;

[0066] S7. After all ants of the generation complete the work node traversal, calculate the path length and the number of turns of the traversal path of each ant according to the pheromone update model, obtain the optimal path of this iteration, that is, the path with the shortest length, and record it, and then update the pheromone;

[0067] S8. Determine whether the iteration number of path search reaches a preset value, if yes, execute the next step; if not, return to S2;

[0068] S9. Calculate and compare the optimal traversal paths of ants of each generation, select the optimal path as the optimal path of the overall search, and save the relevant path information.

[0069] Example 1 (as Figure 2 ):

[0070] Step 1: According to the distribution of obstacles and the location of task points, a grid map is established by combining the grid method, and relevant parameters are set. After debugging, the optimal value of the number of ants m is 50; the number of iterations N maxThe optimal value of is 50; the optimal value of information heuristic factor is 0.6; the optimal value of expected heuristic factor is 5; the optimal value of pheromone concentration Q is 100; the optimal value of pheromone evaporation factor is 0.5; the path length factor adjustment parameter u is 5; the vertical to the track bar operation times factor adjustment parameter v is 20; and the starting point coordinate is (1, 1);

[0071] Step 2, the pheromone concentration between each grid is initialized, and the pheromone between the grids vertical to the track bar direction is reduced, and the reduction coefficient is 0.3, and the pheromone is updated according to the pheromone update formula subsequently;

[0072] Step 3, the same generation of ants is placed at the starting point, and in order to prevent the ants from selecting the walked grid as the target grid, the starting point is first added to the taboo table;

[0073] Step 4, the path search is started, the transition probability of the surrounding untraversed grid is calculated by using the comprehensive heuristic function, then the next traversed grid is selected by using the roulette method, and the grid is added to the taboo table after traversal. If there is no untraversed grid around, that is, the ant is in deadlock, at this time, one of the grids closest to the current grid is selected as the next target node from all the untraversed grids. Considering the running mode of the trolley on the steel mesh, only the upper, lower, left and right grids are selected as the transfer target when transferring to the surrounding grids;

[0074] Step 5, whether the ant traverses all grids is judged, and if yes, the following step is executed; if not, step 4 is returned to continue the path search;

[0075] Step 6, according to the above steps, when all ants of the same generation complete the path search, the optimal path of this iteration is found by considering the path length and the number of times of vertical to the track bar direction running, and the pheromone of each grid of the grid map is updated according to the path and the pheromone update formula;

[0076] Step 7, whether the iteration number of path search reaches the preset value is judged, and if yes, the next step is executed; if not, step 3 is returned;

[0077] Step 8, the optimal paths of each generation of ants are compared, the global optimal path is found, and the related information is saved;

[0078] Step 9, Figure 2 The middle path is the optimal path planned by the application, the obstacle distribution in the map environment is relatively simple, when the ant is in deadlock, the untraversed grid point closest to the point is immediately selected as the next target point, finally an optimal path without collision and with the least number of times of vertical to the track bar direction running is planned, and the feasibility of the application is verified.

[0079] Example two (as Figure 3 ):

[0080] Step 1, according to the obstacle distribution, the task point position, combined with the grid method, the grid map is established, and the related parameters are set. After debugging, the optimal value of the number of ants m is 50; the optimal value of the number of iterations N max The optimal value of the number of iterations N is 50; the optimal value of the information heuristic factor a is 0.6; the optimal value of the expected heuristic factor b is 5; the optimal value of the pheromone concentration Q is 100; the optimal value of the pheromone volatilization factor p is 0.5; the path length factor adjustment parameter u is 5; the vertical to the rail steel operation times factor adjustment parameter v is 20; the starting point coordinates are (1, 1);

[0081] Step 2, initialize the pheromone concentration between each grid, and reduce the pheromone between the grids perpendicular to the rail steel direction, and the reduction coefficient is 0.3, and the pheromone is updated according to the pheromone update formula;

[0082] Step 3, place the ants of the same generation at the starting point, in order to prevent the ants from selecting the traversed grid as the target grid, first add the starting point to the taboo table;

[0083] Step 4, start path search, calculate the transition probability of the surrounding untraversed grid by using the comprehensive heuristic function, then select the next traversed grid by using the roulette method, and add the grid to the taboo table after traversal. If there is no untraversed grid around, that is, the ant is in deadlock, at this time, select one of the other all untraversed grids closest to the current grid as the next target node. Considering the running mode of the trolley on the steel mesh, only the upper, lower, left and right four grids are selected as the transferable target when transferring to the surrounding grids;

[0084] Step 5, judge whether the ant traverses all grids, if yes, execute the following steps; if not, return to step 4 and continue path search;

[0085] Step 6, according to the above steps, when all ants of the same generation complete path search, considering the path length and the number of times of running perpendicular to the rail steel direction, find the optimal path of this iteration, and update the pheromone between the grids of the grid map according to the path and the pheromone update formula;

[0086] Step 7, judge whether the path search iteration number reaches the preset value, if yes, execute the next step; if not, return to step 3;

[0087] Step 8, compare the optimal paths of each generation of ants, find the global optimal path, and save the related information;

[0088] Step 9,Figure 3 The path in the map is the optimal path planned by the application, the map increases the number of obstacles and complicates the distribution of obstacles. When the surrounding untraversed grid points are only left and right grid points during the ant search, the state of the ant is converted into a deadlock state, then the nearest distance between the remaining grid points and the current node is calculated, the distance between the left and right grids is multiplied by the distance coefficient λ when calculating the distance, the grid with the nearest distance is selected as the next target grid point, in this way, when there are untraversed grids in the vertical direction, that is, along the direction of the guide rail reinforcement, the ant will give up the left and right direction grid points and select the untraversed grid points in the up and down direction, finally, an optimal path with no collision and the least running times in the direction perpendicular to the guide rail reinforcement is planned, and the feasibility of the application is verified.

[0089] The application provides a reinforced bar binding robot path planning device based on an improved ant colony algorithm, and specifically comprises the following steps.

[0090] A simulation programming module is used for simplifying a working area into a plurality of task points according to a robot working range, establishing an electronic map according to a grid method, and setting a starting point, an end point and obstacle information of a path.

[0091] An initialization module is used for initializing basic parameters of the ant colony algorithm, determining initial pheromone concentrations between adjacent task points, reducing the pheromone concentrations of adjacent points in the direction perpendicular to the guide rail reinforcement, and substituting the initial pheromone concentrations into a pheromone updating model.

[0092] An initial point determination module is used for generating an ant population, determining the number of the ant population in each iteration, and determining initial working point coordinates.

[0093] A starting iteration module is used for placing each ant at the starting point in the electronic map.

[0094] A first judgment module is used for judging whether there are untraversed nodes around the node where the ant is currently located, if yes, calculating the probability of the ant moving to the next node by using a comprehensive heuristic function, and selecting the next target node by using a roulette method, and if not, expanding the range of the selected node until an untraversed node appears.

[0095] A second judgment module is used for judging whether the current ant has traversed all nodes, if yes, returning to S4, and if not, updating the parent node, returning to S5 and continuing to traverse the working nodes.

[0096] An updating module is used for calculating the path length and the comprehensive index of the turning times of the path traversed by each ant according to the pheromone updating model after all ants in this generation complete the working node traversal, and obtaining the optimal path of this iteration for pheromone updating.

[0097] The third judging module is used for judging whether the iteration number of path searching reaches a preset value, if yes, the next step is executed, and if no, the step S2 is returned to;

[0098] The calculation module is used for calculating and comparing optimal traversal paths of each generation of ants, selecting an optimal path as an optimal path of the whole search, and saving relevant path information.

[0099] With respect to one aspect of the present application, the present application provides a terminal device comprising a memory, a processor and a computer program stored in the memory and capable of running on the processor, the memory stores a computer program capable of running on the processor, and the processor loads and executes the computer program, and adopts the improved ant colony algorithm-based path planning method of a steel bar binding robot.

[0100] In summary, the present application proposes an improved ant colony algorithm-based path planning method of a steel bar binding robot in view of the operation characteristics of the steel bar binding robot driving along the guide rail steel bar and jacking and sliding perpendicularly to the guide rail steel bar, and considering reducing the jacking times to complete the traversal of all operation nodes, the method firstly determines the coordinates of all binding points and obstacles in the steel bar mesh operation area, then determines a circular operation area according to the operation range of the mechanical arm of the binding robot, and then covers all the binding points with the least circles, takes the center of each circle as an operation task point of the binding robot, and converts the obstacle region formed by the irregular-shaped obstacles into a regular pattern in a cutting manner, and converts the operation environment into a grid electronic map in combination with the distribution of the task points and the obstacle region; in the process of ant searching a path, the improved heuristic function and pheromone update formula can reduce the probability of running in the direction perpendicular to the guide rail steel bar in the case that there is an untraversed operation point along the guide rail direction, and preferentially considers the operation points in the guide rail direction, so as to reduce the jacking and sliding times in the final traversal path perpendicular to the guide rail steel bar; when the ant falls into a deadlock and there is no untraversed point around the neighboring nodes, the search range is expanded to find the nearest untraversed operation point as the next node, so as to solve the deadlock problem.

[0101] It should be noted that, in the present document, the terms such as first and second, etc., are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device.

[0102] Those of ordinary skill in the art, with the benefit of this description, can appreciate the disclosure in conjunction with its specific features, embodiments, and examples. The above-described features and specific embodiments are to be considered illustrative of specific ways to make and use the disclosure, and such to provide a concrete basis to enabling the patent. Various modifications, both as to the specific nature of the features and embodiments and methods of their implementation, can be apparent from the description. The scope of the disclosure is defined by the appended claims and their equivalents.

[0103] While the embodiments of the present disclosure have been illustrated and described, it will be understood by those skilled in the art that various changes, modifications, alternatives, and variations can be made therein without departing from the spirit and scope of the present disclosure, which is defined by the appended claims and their equivalents.

[0104] In the description of the specification, reference to "one embodiment", "an example", "a specific example", and so on, means that a particular feature, structure, material, or characteristic being described is included in at least one embodiment or example of the disclosure. The appearances of the above-described terms in various places in the specification are not necessarily referring to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

Claims

1. A method for path planning of a steel binding robot based on an improved ant colony algorithm, characterized in that, The specific steps include: S1. According to the robot operation range, the working area is simplified into several operation nodes, an electronic map is established according to the grid method, and the starting point, end point and obstacle information of the path are set; S2. The parameters of the ant colony algorithm are initialized, the initial pheromone concentration between adjacent operation nodes is determined, the pheromone concentration of adjacent points perpendicular to the direction of the guide rail reinforcement is reduced, and the initial pheromone concentration is substituted into the pheromone update model, wherein the parameters of the ant colony algorithm are: the number of ants m, the maximum iteration number Nmax, the information heuristic factor α, the expected heuristic factor β, the pheromone volatilization factor ρ, the pheromone concentration Q, and the starting point coordinates of the robot; S3. Generate an ant population, determine the number of ant populations for each iteration, and determine the initial operation point coordinates; S4. Put each ant into the starting point in the electronic map; S5. Determine whether there are untraversed nodes around the current node of the ant, if so, calculate the probability of the ant moving to the next node using the comprehensive heuristic function, and select the next target node using the roulette method; if not, expand the range of selected nodes until an untraversed node appears; The expression of the comprehensive heuristic function in S5 is as follows: In the formula, is a comprehensive heuristic function, representing the expectation of the kth ant transferring from the i node to the j node; is the distance between the i node and the j node; is a distance coefficient, taking 1 when the i node and the j node are in the same direction of the guide rail reinforcement, and taking a value greater than 1 when the i node and the j node are not in the same direction of the guide rail reinforcement, and the value is taken according to requirements. S6. Determine whether the current ant has traversed all nodes, if so, return to S4; if not, update the current node of the ant to the next target node determined in S5, return to S5, and continue to traverse the operation nodes; S7. After all ants of this generation complete the operation node traversal, calculate the path length and the comprehensive index of the number of turns of the path traversed by each ant according to the pheromone update model, obtain the optimal path of this iteration, that is, the path with the shortest length, and record it down, and then update the pheromone; The expression of the pheromone update model is: ; In the formula, represents the pheromone on the path from job point i to j at time t, is the evaporation rate of pheromone, is the pheromone content before the path, is the pheromone increment of the kth ant on the path (i,j) at time t, Q is the pheromone concentration, is the comprehensive index of the pheromone on the path walked by the kth ant in the tth iteration, the smaller the index, the better the path, and the more pheromone left, is the length of the path, is the number of times of moving along the direction perpendicular to the track reinforcement, u and v are adjustment parameters of the two factors, and appropriate values can be taken according to needs; S8. Determine whether the iteration number of path search reaches the preset value, if so, execute the next step; if not, return to S2; S9. Calculate and compare the optimal traversal paths of ants of each generation, select the optimal path as the optimal path of this overall search, and save the related path information.

2. The improved ant colony algorithm-based path planning method for a steel bar tying robot according to claim 1, characterized in that: The transfer probability formula of the ant moving to the next node in S5 is: In the formula, C is the feasible region set of the next node of the ant; is the pheromone value of path (i, j); a is the pheromone heuristic factor, which determines the influence degree of pheromone concentration on path selection; β is a desired heuristic factor, which determines the degree of influence of heuristic information on path selection, P (i, S, t) represents the pheromone value of path (i, S) at time t, S belongs to a selectable node in C; P (i, S, t) is the expectation of path (i, S) at time t.

3. The improved ant colony algorithm-based path planning method for a steel bar tying robot according to claim 2, characterized in that, According to the robot operation range, the binding points on the reinforcement mesh of the operation area are covered with the least number of circles, and the center of each circle is taken as the task point of the robot operation.

4. The improved ant colony algorithm-based path planning method for a steel bar tying robot according to claim 3, characterized in that, The obstacle area formed by the irregularly shaped obstacles is converted into a regular pattern by cutting, and the operation environment is converted into a grid map in combination with the distribution of the task points and the obstacle area.

5. The improved ant colony algorithm-based path planning method for a steel bar tying robot according to claim 1, characterized in that: The electronic map is established by using a simulation software, which is MATLAB R2022b.

6. The device for path planning of a steel bar binding robot based on an improved ant colony algorithm, characterized in that, It comprises: A simulation programming module is used to simplify the working area into several task points according to the robot operation range, establish an electronic map according to the grid method, and set the starting point, end point and obstacle information of the path; The initialization module is configured to initialize parameters of the ant colony algorithm, determine initial pheromone concentrations between adjacent job nodes, reduce the pheromone concentrations of adjacent points perpendicular to the direction of the rail reinforcement, and substitute the initial pheromone concentrations into a pheromone update model, wherein the parameters of the ant colony algorithm include the number of ants m, the maximum number of iterations Nmax, the information heuristic factor a, the expected heuristic factor b, the pheromone evaporation factor p, the pheromone concentration Q, and the starting point coordinates of the robot. The initial point determination module is configured to generate an ant population, determine the number of ants in each iteration, and determine initial job point coordinates. The starting iteration module is configured to place each ant at the starting point in the electronic map. The first judgment module is configured to determine whether there are untraversed nodes around the current node of the ant, and if so, calculate the probability of the ant moving to the next node by using a comprehensive heuristic function, and select the next target node by using the roulette method; if not, expand the range of the selected node until an untraversed node appears. The expression of the comprehensive heuristic function is as follows: In the formula, is a comprehensive heuristic function, representing the expectation of the kth ant transferring from the i node to the j node; is the distance between the i node and the j node; is a distance coefficient, taking 1 when the i node and the j node are in the same direction of the guide rail reinforcement, and taking a value greater than 1 when the i node and the j node are not in the same direction of the guide rail reinforcement, and the value is taken according to requirements. The second judgment module is configured to determine whether the current ant has traversed all nodes, and if so, return to S4; if not, update the parent node and return to S5 to continue the job node traversal. The update module is configured to, after all ants in the generation complete the job node traversal, calculate the path length and the comprehensive index of the number of turns of the traversal path of each ant according to the pheromone update model, and obtain the optimal path of the current iteration for pheromone update. The expression of the pheromone update model is as follows: ; In the formula, represents the pheromone on the path from job point i to j at time t, is the evaporation rate of pheromone, is the pheromone content before the path, is the pheromone increment of the kth ant on the path (i,j) at time t, Q is the pheromone concentration, is the comprehensive index of the pheromone on the path walked by the kth ant in the tth iteration, the smaller the index, the better the path, and the more pheromone left, is the length of the path, is the number of times of moving along the direction perpendicular to the track reinforcement, u and v are adjustment parameters of the two factors, and appropriate values can be taken according to needs; The third judgment module is configured to determine whether the number of path search iterations reaches a preset value, and if so, execute the next step; if not, return to S2. The calculation module is configured to calculate and compare the optimal traversal paths of the ants in each generation, select the optimal path as the optimal path of the current overall search, and save the relevant path information.

7. A terminal device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, The memory stores a computer program capable of running on the processor, and the processor loads and executes the computer program, thereby adopting the reinforcement binding robot path planning method based on the improved ant colony algorithm according to any one of claims 1 to 5.

Images