Material sorting mechanical arm path planning method and system based on improved ant colony algorithm

Abstract

The present application relates to the technical field of path planning, and especially relates to a material sorting mechanical arm path planning method and system based on an improved ant colony algorithm, which comprises the following steps: constructing a three-dimensional space plane of a mechanical arm, extracting grid points required for a three-dimensional path from the three-dimensional space by using a space equal division method; modeling threats and terrain obstacles in an environment; constructing a target function in a mechanical arm operation process; constructing a preliminary checkpoint set and calculating the number of cross sections; obtaining a theoretical shortest path of the mechanical arm according to a starting point and an ending point of a mechanical arm trajectory path; rounding an intersection point of the theoretical shortest path and a cross section composed of a Y-axis and a Z-axis, and constructing a plurality of rectangular planes with the rounded intersection point as a center; uniformly selecting a plurality of preliminary checkpoints in each rectangular range; and constructing an improved ant algorithm to optimize the mechanical arm trajectory path. The present application solves the problems of a long operation trajectory of a mechanical arm and poor path optimization performance in the prior art.

Description

Technical Field

[0001] This invention relates to the field of path planning technology, and in particular to a path planning method and system for a material sorting robotic arm based on an improved ant colony algorithm. Background Technology

[0002] Robotic arms are playing an increasingly important role in modern industrial production and daily life, attracting widespread attention from researchers both at home and abroad. Research hotspots mainly focus on the motion planning and control of robotic arms, among which trajectory planning is the foundation of robotic arm trajectory control and is of great significance to the stability, operating efficiency, operational accuracy and energy consumption of robotic arms.

[0003] The path planning problem for robotic arms is common in smart warehousing and smart factories. Its core task is to plan a suitable trajectory for the end point of the robotic arm from the starting point to the target point, based on the known origin and destination points of the material handling and the workspace environment. The planned trajectory must take into account the stability and safety of the material handling device, obstacle avoidance, and ensure the stable operation of the robotic arm itself, while also meeting the performance constraints of the robotic arm. In addition, the trajectory also needs to be optimal under a certain performance metric to maximize the working efficiency of the robotic arm. Therefore, the path planning problem for robotic arms is essentially an optimization problem that seeks the optimal solution.

[0004] There are two types of constraints when solving this problem: First, various devices in the working environment hinder the operation of the robotic arm, and the specific location, length, shape and other factors of the devices need to be considered; second, the performance of the robotic arm itself also has a series of constraints, such as the limitations on the operating speed and acceleration of each joint of the robotic arm, the limitations on the rotation angle of each axis, the magnitude of torque and the size of the load on the robotic arm, etc.

[0005] In summary, the professionalism of the material sorting robot path planning problem lies in the fact that it requires ensuring the accurate transport of identified materials to the correct location while also considering the stability and safety of the robot itself, and minimizing the robot's travel trajectory as much as possible. Solving this problem can improve sorting efficiency, reduce labor costs, and is applicable to various types of material handling tasks across different industries. Summary of the Invention

[0006] To address the shortcomings of existing methods, this invention solves the problem that the robotic arm's running trajectory is too long and the path optimization performance is poor in existing methods.

[0007] The technical solution adopted in this invention is: a path planning method and system for a material sorting robotic arm based on an improved ant colony algorithm, comprising the following steps:

[0008] Step 1: Construct the three-dimensional space plane of the robotic arm and extract the grid points required for the three-dimensional path from the three-dimensional space using the equal division space method;

[0009] Step 2: Model the threats and terrain obstacles in the robotic arm's environment;

[0010] Step 3: Construct the objective function for the robotic arm's operation;

[0011] Step 4: Construct a preliminary checkpoint set and calculate the number of cut surfaces;

[0012] Furthermore, the formula for the number of facets is:

[0013]

[0014] Among them, L sg (x) represents the distance between the starting point and the ending point along the X-axis, S w The width of the cut surface.

[0015] Setting the cross-sectional width to an adjustable parameter reduces the number of acceleration changes at the robotic arm's joints, making the robotic arm's operation more stable.

[0016] Step 5: Obtain the theoretical shortest path of the robotic arm based on the starting and ending points of the robotic arm trajectory; take the intersection of the theoretical shortest path and the tangent formed by any two axes, round down the points, and construct several rectangular ranges centered on the rounded intersection points. Select several pre-check points evenly within each rectangular range.

[0017] Furthermore, checkpoints that are duplicates or located outside the map area are removed.

[0018] Compared to existing methods that uniformly select points across the entire cross-section, this method uses different rectangles to select points, which, while retaining global exploration capabilities, balances refinement and accuracy, significantly increasing the probability of the algorithm obtaining a better path.

[0019] Step 6: Construct an improved ant algorithm to optimize the trajectory path of the robotic arm.

[0020] Furthermore, specifically including:

[0021] Step 61: Initialize ant colony parameters and calculate the heuristic value between each checkpoint and the next layer checkpoint;

[0022] Furthermore, the formula for calculating the heuristic value is:

[0023]

[0024] Among them, (x i ,y i ,z i(x) represents the coordinates of the current checkpoint i, and (x) represents the coordinates of the current checkpoint i. j ,y j ,z j ) represents the coordinates of the next checkpoint j.

[0025] Step 62: Ants search for feasible paths. Starting from the origin, the ants choose the next waypoint using a roulette wheel method.

[0026] Step 63: Check the obtained path and update the local pheromone to obtain the pheromone change value of the checked path; fit the waypoints obtained by the ants into the trajectory of the robotic arm, and calculate the motion trajectory path of the end effector and the change of angle of each joint.

[0027] Furthermore, the formula for obtaining the pheromone change value of the inspected path is:

[0028]

[0029] In the formula, iter is the current iteration number, t is the number of times the search segment has been searched, iter_max is the maximum iteration number, and the local pheromone intensity Q is a constant; L path (tk) represents the length of the current entire path; △τ i,j (t) is the old pheromone change value when moving from grid point i to grid point j.

[0030] Step 64: Improve the pheromone evaporation factor of the current iteration pheromone change value;

[0031] Furthermore, the improved formula for the pheromone volatile factor is as follows:

[0032]

[0033] In the formula, Represents the global evaporation constraint coefficient; C is the compensation coefficient, iter is the current iteration number, and ρ(iter) represents the pheromone evaporation factor at the iter-th iteration. min This represents the minimum value of the pheromone evaporation factor.

[0034] The beneficial effects of this invention are:

[0035] 1. Ant colony algorithm can select checkpoints. Considering the working environment of the material sorting robot arm, after determining the starting and ending points of the handling, the working space of the robot arm is divided along the X-axis, generating a series of cross-sections between the starting and ending points. Then, a straight line is constructed based on the positions of the starting and ending points. This straight line intersects with each cross-section. A rectangular range is constructed with the intersection point as the center, and preliminary checkpoints are evenly selected within the rectangular range. Compared with the original method of evenly selecting points on the entire cross-section, this point selection method retains the global exploration capability while taking into account refinement and accuracy, which greatly improves the probability of the algorithm obtaining a better path.

[0036] 2. Improved Ant Colony Pheromone Update: The pheromone update method of the ant colony algorithm is improved. In local updates, for paths explored by ants, the path segments are checked, and different pheromone update methods are set for impassable path segments, previously searched path segments, and previously unsearched path segments. For impassable path segments, the pheromone value is set to -1, and since the initial pheromone value is 1, the probability of it being selected in the next iteration is 0. For searched path segments, the pheromone is updated according to the length of the currently searched path. For newly searched path segments, more pheromone is added at the beginning to increase the probability of it being selected in the next iteration. As the number of iterations increases, the pheromone obtained by newly searched path segments gradually decreases, which can effectively control and adjust environmental factors.

[0037] 3. When updating the global pheromone, the pheromone evaporation coefficient is made adaptive, changing with the number of iterations. The adjustment strategy of the global pheromone evaporation coefficient can effectively control and optimize system performance, better overcome local optima, and thus explore the global optimal solution more effectively. Attached Figure Description

[0038] Figure 1 This is a detailed diagram of the material sorting robotic arm path planning method and system based on the improved ant colony algorithm of the present invention;

[0039] Figure 2 This is a schematic diagram of the three-dimensional grid map of the present invention;

[0040] Figure 3 This is a three-dimensional simulation diagram of the threat equivalent terrain of the present invention;

[0041] Figure 4 This is a comparison diagram of the optimal path between the method of this invention and the ACA algorithm;

[0042] Figure 5 This is a comparison chart of the optimal path convergence between the method of this invention and the ACA algorithm;

[0043] Figure 6 It is a rectangular diagram constructed based on the intersection of the cross-section and the shortest path. Detailed Implementation

[0044] The present invention will be further described below with reference to the accompanying drawings and embodiments. The drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0045] like Figure 1 As shown, a path planning method and system for a material sorting robotic arm based on an improved ant colony algorithm includes the following steps:

[0046] Step 1: Construct the three-dimensional spatial plane of the robotic arm's workspace, and use the equal division spatial method to extract the grid points required for the three-dimensional path from the three-dimensional space;

[0047] like Figure 2 As shown, the workspace of the robotic arm is determined by taking the vertex at the lower left corner of the 3D map as the origin A of the 3D space coordinate system. A 3D coordinate system is established at point A, where the X-axis is along the direction of increasing longitude, the Y-axis is along the direction of increasing latitude, and the Z-axis is perpendicular to the sea level.

[0048] Taking point A as the vertex, take the maximum length of the 3D map along the X-axis as AB = n, along the Y-axis as AD = m, and along the Z-axis as AA1 = h. This constructs a cubic region ABCD-A1B1C1D1 containing the 3D map, which is the planning space for the 3D path.

[0049] After the 3D path space is established, the grid points required for the 3D path are extracted from the 3D space using the method of equal space division. First, the space ABCD-A1B1C1D1 is divided into n equal parts along the X-axis side AB, resulting in n+1 YZ planes. Second, these n+1 planes are divided into m equal parts along the Y-axis side AD, resulting in m+1 XZ planes. Then, the space is divided into h equal parts along the Z-axis side AA1, resulting in h+1 XY planes. Finally, the intersection points are calculated to obtain the grid points used for ant colony algorithm planning.

[0050] The intersections of these planes in three-dimensional space are the grid points used in the ant colony algorithm; since the workspace is not necessarily a cube, a generalized length setting is adopted.

[0051] Step 2: Model the threats and terrain obstacles in the environment;

[0052] After constructing the three-dimensional path space, the threats and terrain obstacles in the environment are modeled to obtain a mathematical model of threat-equivalent terrain:

[0053]

[0054] In the formula, z(x,y) is the terrain height function, x and y are the abscissa and ordinate of a point on the horizontal projection plane, respectively, and z is the terrain height corresponding to the coordinate point (x,y) on the horizontal projection plane. Represents the center coordinates of the j-th peak; and These are the attenuation amounts of the j-th peak along the X and Y axes, respectively, controlling the slope; h0 is the baseline elevation of the terrain, representing the ground elevation when there are no threats or obstacles; N is the number of threats and obstacles. The maximum height of the j-th threat or obstacle; and It is a parameter used to control the range of influence of threats or obstacles; T j Let be the range of influence on the horizontal projection surface of the j-th threat or obstacle.

[0055] like Figure 3 As shown, by changing these parameters in the mathematical model of threat-equivalent terrain, it is possible to simulate threat-equivalent mountain terrain and realize mountains with different numbers, heights and slopes.

[0056] Step 3: Construct the fitness value of the objective function during the operation of the robotic arm;

[0057] After adopting the threat-equivalent terrain model, let P be the starting point of the robotic arm's trajectory path. s (x s ,y s ,z s The endpoint is P. g (x g ,y g ,z g The starting and ending points must not be obstacle points. The raster data of the threat-equivalent terrain in the raster map is obtained to determine passable raster points. Furthermore, the robotic arm should minimize the cumulative joint rotation during operation to reduce energy consumption, and for safety and stability, it must not collide with obstacles. Based on these requirements, an optimized objective function fitness value f is constructed. K The objective function is defined by the obstacle avoidance objective f. CO The shortest motion path f of the robotic arm end effector l The objective f is to minimize the change in the angle of each joint. Q The composition, as shown in the formula below:

[0058] f K =f CO *(η1f Q +η2f l (2)

[0059] Where η1 and v2 are f Q f lThe weighting coefficient, f CO It is the obstacle avoidance target; when a collision occurs, f CO =inf, when no collision occurs, f CO =1.

[0060] Step 4: Construct the preliminary checkpoint set `allow` and initialize the parameters; between the start and end points, according to the set cross-sectional width `S` along the X-axis. w Calculate the number of cut surfaces S n Number of cross-sections S n The calculation method is as shown in equation (3):

[0061]

[0062] Among them, L sg (x) is the starting point P s (x s ,y s ,z s ) and endpoint P g (x g ,y g ,z g The distance along the X-axis, i.e., L sg (x)=|x s -x g |,S w The width of the cut surface.

[0063] The cross-sectional width S of this invention w It is set according to the performance of the machine, for example, setting S. w =20, the robotic arm's running path consists of multiple straight lines no shorter than 20. This allows the robotic arm to have a longer straight running path during operation, the joint speed will not change drastically, the joint acceleration will not change frequently, and the robotic arm will run more stably. This is the first step of the ant colony algorithm. According to the working conditions of the robotic arm, the robotic arm moves the object from side a to side b. The direction of this ab movement is aligned with the X-axis direction to perform a tangential operation.

[0064] Step 5: Based on the starting point P of the robotic arm's trajectory path s (x s ,y s ,z s ) and endpoint P g (x g ,y g ,z g The location determines the theoretical shortest path L. small The equation of the line containing the given information is as follows:

[0065]

[0066] Among them, (x s ,y s ,z s (x) represents the coordinates of the starting point s of the path, and (x) represents the coordinates of the starting point s of the path. g ,y g ,z g ) represents the coordinates of the path endpoint g, and the parameter t∈[0,1], where t=0 represents the coordinates of the starting point and t=1 represents the coordinates of the endpoint.

[0067] like Figure 6 Taking a cross-section formed by the Y-axis and Z-axis as an example (it could also be the XY-axis or XZ-axis), the shortest path L small The intersection points with each cross-section are (x, y, z). The existing method rounds the intersection points and uniformly selects 100 grid checkpoints on each cross-section. This invention constructs three rectangular ranges centered on the intersection points on each cross-section, i.e., the first to third range outlines in the figure. The area of ​​the rectangle is determined according to the influence range of the obstacle in equation (1). It is easy to see that the maximum influence range of a single obstacle is... The experiment was set up The side length of the rectangle inscribed in this range is The ant colony is a square with an circumscribed rectangle of side length 20. A rectangle with the side length of the inscribed rectangle is used to construct the first area; a rectangle with the side length of the circumscribed rectangle is used to construct the second area; and a rectangle with the side length of the circumscribed rectangle is used to construct the third area. 36, 36, and 25 grid points are selected as preliminary checkpoints within the first, third, and fourth area boundaries, respectively. Selecting points within the first area increases the number of points near the intersection of the shortest path and the tangent plane, increasing the probability that the ant colony's search path is closer to the shortest path and significantly improving path quality. Simultaneously, the second and third areas increase in size and select points separately, ensuring that the entire area of ​​selected points overlaps as much as possible with the entire tangent plane area, guaranteeing the ability to explore the tangent space and preventing the selection of points from being blocked by obstacles, thus preventing the search for feasible paths. Duplicate points and points outside the map area need to be removed. Passable checkpoints are checked based on map information and saved to the preliminary checkpoint set `allow`. The above areas and the number of checkpoints can be customized according to specific circumstances.

[0068] Existing ant colony algorithms construct preliminary checkpoints by uniformly selecting them across the entire cross-section at a certain step size. To enhance the purposefulness of checkpoint selection, this invention selects checkpoints around the intersection of the straight line formed from the start point to the end point and the cross-section, ensuring the direction and purposefulness of the ant colony algorithm path.

[0069] Step 6: Construct an improved ant colony algorithm to optimize the trajectory path of the robotic arm;

[0070] Step 61: Initialize ant colony parameters, including: number of ants m, maximum number of iterations iter_max, pheromone factor α, heuristic function factor β, pheromone evaporation factor ρ, and local pheromone intensity Q.

[0071] Initialize the pheromone value τ between each checkpoint and the next checkpoint to 1, and set the coordinates of the current point to P. i (x i ,y i ,z i The coordinates of the next checkpoint are P. j (x j ,y j ,z j ), and initialize the heuristic value η between each checkpoint and the next layer checkpoint according to formula (5). ij The formula is:

[0072]

[0073] Among them, (x i ,y i ,z i (x) represents the coordinates of the current checkpoint i, and (x) represents the coordinates of the current checkpoint i. j ,y j ,z j ) represents the coordinates of the next checkpoint j.

[0074] Because the method of selecting checkpoints has changed and is more purposeful, when selecting checkpoints for a new layer of cross-sections, it is only necessary to ensure that the path between two adjacent layers is as short as possible. Therefore, the calculation method of the heuristic value has been changed from being determined by the distance between the current point and the endpoint to being determined by the distance between the current point and the checkpoint of the next layer.

[0075] Step 62: Ants search for feasible paths. Starting from the origin, the ants choose the next waypoint using a roulette wheel approach. ij It is the probability of moving from point i to point j, calculated according to formula (6).

[0076]

[0077] Where α is the pheromone factor, β is the heuristic function factor, and η ik τ(t) represents the heuristic value for an ant to move from grid point i to the next layer checkpoint k. ik (t) represents the pheromone value when moving from i to the next checkpoint k.

[0078] η ij (t) represents the heuristic value for the ant to move from grid point i to target point j, τ ij(t) represents the pheromone value from i to j, and allow(i+1) represents the set of passable grid points in the next layer; the ant searches along the cut surface sequentially until it reaches the destination.

[0079] Step 63: Inspect the obtained path and update the local pheromone to obtain the pheromone change value of the inspected path; fit the waypoints obtained by the ants into the trajectory of the robotic arm, and calculate the motion trajectory path f of the end effector. L and the change in angle of each joint f Q Perform collision detection on the path to obtain the collision target f. CO The value of is recorded, and the path segment where the collision occurred is recorded; based on the inspection results, the path segment on the path is locally updated according to formula (7), and the update is performed according to whether the path segment is passable and whether the path segment has been explored in the previous ant search. The formula is:

[0080]

[0081] In the formula, iter is the current iteration number, and the local pheromone intensity Q is a constant; L path (tk) represents the length of the current entire path; △τ i,j (t+1) is the new pheromone change value from grid point i to grid point j; Δτ i,j (t) is the old pheromone change value when moving from grid point i to grid point j.

[0082] The pheromone update method of this invention involves updating the pheromone after each ant has explored the path. This pheromone update reduces the probability of the retrieved path being selected, thus preventing the ant from getting trapped in a local optimum. The new pheromone update method is offline, meaning that in the same iteration, the probability of all ants selecting a new point is not affected by other ants. Furthermore, the pheromone update method is changed according to the path conditions. If the path is impassable, a negative one is taken, causing the pheromone level to be set to zero during the global pheromone update. When a new path is obtained, a larger amount of pheromone is accumulated, and this amount approaches that of the old path as the number of iterations increases. When an old path is obtained again, the accumulated pheromone level is increased based on the original pheromone level.

[0083] Step 64: Extract the pheromone change values ​​of each road segment in the local pheromone update, and perform a global pheromone update for all road segments; after one iteration, update the pheromone of all road segments according to formula (8). The pheromone evaporation factor ρ will change with the number of iterations, as shown in formula (9). This iteration ends.

[0084] τ i,j (iter+1)=(1-ρ)*τ i,j (iter)+△τ i,j(iter) (8)

[0085]

[0086] In the formula, ρ represents the global evaporation constraint coefficient; C is the compensation coefficient, used to keep the evaporation coefficient within a fixed range; iter is the current iteration number, and ρ(iter) represents the pheromone evaporation factor at the iter-th iteration. min This represents the minimum value of the pheromone evaporation factor.

[0087] In the original ant colony algorithm's global pheromone update method, the evaporation factor ρ is fixed. The modified coefficient gradually decreases as the number of iterations increases. This modification causes the proportion of pheromones evaporating on the path to gradually decrease as the iterations proceed. In the early stages of the algorithm iteration, this increases the probability of discovering new paths, and in the later stages of the algorithm iteration, it accelerates the convergence speed of the algorithm.

[0088] The optimization problem is evaluated to find the globally optimal path.

[0089] The theoretical shortest path L has been obtained. small After one iteration, the optimal fitness value for this iteration is determined based on the fitness values ​​of the paths of m ants. The fitness value of the paths searched by each ant is calculated according to formula (2). The global optimal path for this iteration is found based on the fitness value. It is determined whether the path has reached the theoretical shortest path or whether the number of iterations has reached the predetermined maximum number of iterations. This step is to confirm whether further optimization is needed. If either of these conditions is met, the system will output the current optimal path. However, if the above conditions are not met, the system will jump to step 62 and continue the subsequent optimization iteration process. This iteration process will continue until the theoretical shortest path is reached or the maximum number of iterations is reached, so as to ensure that the most optimized path solution is finally obtained.

[0090] To verify the effectiveness and superiority of the Ant Colony Algorithm (ACA) and the Improved Ant Colony Algorithm (IACA) in robotic arm trajectory planning, this invention conducted simulation experiments based on the Ant Colony Algorithm and the Improved Ant Colony Algorithm, and analyzed the results.

[0091] Simulation environment settings: The simulation platform is MATLAB 2021b, the processor is Intel(R) Core(TM) i5-8300H CPU@2.30GHz, and the memory is 16GB.

[0092] Equivalent terrain of the robotic arm's workspace, such as Figure 4As shown, the workspace of the robotic arm's trajectory is set to a 100cm*100cm*100cm area, with the starting point coordinates being (18,16,10) and the ending point coordinates being (93,79,50). Mountain terrain is used as an equivalent of obstacles and accumulated materials. Ant colony algorithm parameters are set as follows: number of ants m = 10, maximum number of iterations iter_max = 100, pheromone factor α = 5, heuristic function factor β = 1, pheromone evaporation factor ρ = 0.1, and local pheromone intensity Q = 1.

[0093] Figure 4 This is a comparison of the optimal paths obtained by the two algorithms. The squares represent the starting points, and the stars represent the ending points. The solid lines represent the paths obtained by the IACA algorithm, and the dashed lines represent the paths obtained by the ACA algorithm. Observing the drawn paths, it's noticeable that the improved path is smoother and closer to a straight line. This indicates that the improved algorithm more fully considers the terrain characteristics during path planning, resulting in smoother operation and a significantly shorter path length.

[0094] A line graph showing the relationship between the number of iterations and the fitness value is shown below. Figure 5 As shown, observing the trend of the curve, the improved algorithm IACA is able to find a shorter path in the initial stage, which is due to the adoption of an innovative point selection scheme. Subsequently, the curve drops rapidly and tends to stabilize. In addition, comparing the final fitness values ​​of the two algorithms under the same number of iterations, the improved algorithm performs better in terms of fitness value, which means that it has a greater advantage in problem solving.

[0095] The simulation results show the theoretical and actual optimal fitness values ​​as shown in Table 1. The improved ant colony algorithm has a relatively small difference error between the theoretical and actual optimal fitness values, which indicates that the improved ant colony algorithm has achieved significant optimization effect on the robotic arm path planning problem and can effectively approach the expected optimal solution. The algorithm fully considers the influence of terrain and obstacles in trajectory planning, making the robotic arm's running path more reasonable and safe.

[0096] Table 1. Simulation results: theoretical optimal fitness and actual optimal fitness.

[0097]

[0098] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A path planning method for a material sorting robotic arm based on an improved ant colony algorithm, characterized in that, Includes the following steps: Step 1: Construct the three-dimensional space plane of the robotic arm and extract the grid points required for the three-dimensional path from the three-dimensional space using the equal division space method; Step 2: Model the threats and terrain obstacles in the robotic arm's environment; Step 3: Construct the objective function for the robotic arm's operation; Step 4: Construct a preliminary checkpoint set and calculate the number of cut surfaces; The formula for the number of facets is: ; in, This represents the distance between the starting point and the ending point along the X-axis. The width of the cut surface; Step 5: Obtain the theoretical shortest path of the robotic arm based on the starting and ending points of the robotic arm trajectory; take any two axes of X, Y, and Z as the tangent plane, round the intersection point of the theoretical shortest path and the tangent plane, and construct several rectangular ranges with the rounded intersection points as the center, and uniformly select several pre-check points within each rectangular range. Step 6: Construct an improved ant colony algorithm to optimize the trajectory path of the robotic arm; Step six specifically includes: Step 61: Initialize ant colony parameters and calculate the heuristic value between each checkpoint and the next layer checkpoint; Step 62: Ants search for feasible paths. Starting from the origin, the ants choose the next waypoint using a roulette wheel method. Step 63: Check the obtained path and update the local pheromone to obtain the pheromone change value of the checked path; fit the waypoints obtained by the ants into the trajectory of the robotic arm, and calculate the motion trajectory path of the end effector and the change of angle of each joint. The formula for obtaining the pheromone change value of the inspected path is: ； In the formula, iter This represents the current iteration number. t To retrieve the number of times a road segment has been searched, iter_max The maximum number of iterations, and the local pheromone intensity. Q It is a constant; Indicates the length of the current entire path; From grid points i Transfer to grid point j The old pheromone change value; Step 64: Improve the pheromone evaporation factor of the current iteration pheromone change value; The improved formula for pheromone volatile factor is: ； In the formula, This represents the global volatile constraint coefficient; For compensation coefficient, iter This represents the current iteration number. Indicates the first iter The pheromone evaporation factor at the next iteration This represents the minimum value of the pheromone evaporation factor.

2. The material sorting robotic arm path planning method based on the improved ant colony algorithm according to claim 1, characterized in that, In step five, checkpoints that are duplicates or located outside the map area are removed.

3. The material sorting robotic arm path planning method based on the improved ant colony algorithm according to claim 1, characterized in that, The formula for calculating the heuristic value is: ； in, Current checkpoint coordinates, For the next level of checkpoint j The coordinate values.

4. A path planning system for a material sorting robotic arm based on an improved ant colony algorithm, characterized in that, include: Memory is used to store instructions that can be executed by the processor; A processor for executing instructions to implement the material sorting robotic arm path planning method based on the improved ant colony algorithm as described in any one of claims 1-3.

5. A computer-readable medium storing computer program code, characterized in that, The computer program code, when executed by a processor, implements the material sorting robotic arm path planning method based on the improved ant colony algorithm as described in any one of claims 1-3.

Images