A mobile robot path planning method based on improved ant colony algorithm

Abstract

This invention belongs to the field of robot path planning and aims to solve the problem of achieving better path planning in complex environments. This invention provides a mobile robot path planning method based on an improved ant colony algorithm. This method first establishes environmental grid map information and determines the starting and ending points for each mobile robot. It then uses the improved ant colony algorithm to plan a global path for each mobile robot and performs targeted secondary optimization. This results in a simpler, smoother, and shorter output path, providing an effective solution for multi-robot path planning in complex environments.

Description

Technical Field

[0001] The present invention belongs to the field of robot path planning, and in particular relates to a mobile robot path planning method based on an improved ant colony algorithm. Background Art

[0002] Mobile robot path planning algorithms are an important field in robotics, which involve how to enable a robot to move autonomously from one location to another in an environment while avoiding collisions with obstacles during movement.

[0003] With broad application prospects in autonomous driving, industrial automation, service robotics, and other fields, the research and development of mobile robot path planning algorithms is crucial for improving the autonomy, flexibility, and efficiency of robots. These algorithms typically need to consider multiple factors, such as the dynamic changes in the environment, the dynamic characteristics of the robot, task priorities, and energy consumption.

[0004] Traditional mobile robot path planning algorithms work well in simple path planning applications and are widely used in simple mobile robot path planning problems such as single drones and single unmanned vehicles. However, when faced with complex environments, traditional path planning algorithms are difficult to achieve ideal results. Therefore, applying heuristic intelligent algorithms with learning capabilities to path planning is a major current trend; most intelligent optimization algorithms are based on certain rules summarized from group behavior in nature, and swarm intelligence algorithms are proposed based on these rules; this type of algorithm has good solving capabilities for high-dimensional, complex, and multi-constrained optimization problems; common intelligent optimization algorithms include ant colony algorithms, particle swarm algorithms, genetic algorithms, fish swarm algorithms, etc.

[0005] The ant colony algorithm (ACO) is an intelligent biomimetic algorithm that simulates the foraging behavior of ant colonies in nature. Its core concept is to continuously update the pheromone concentration along a path through positive feedback, until the pheromone concentration on the optimal path is maximized, thereby finding the optimal solution. Due to its positive feedback, parallel computing, and robustness, the ACO has achieved good results in mobile robot path planning. However, the classic ACO still has many shortcomings. Therefore, improving the ACO is particularly important to ensure the efficient completion of mobile robot path planning. Summary of the Invention

[0006] In order to solve at least one of the above technical problems existing in the prior art, the present invention provides a mobile robot path planning method based on an improved ant colony algorithm.

[0007] The present invention is implemented by the following technical solution: a mobile robot path planning method based on an improved ant colony algorithm, comprising the following steps:

[0008] S1: Establishing environmental grid map information, and determining the movement starting point and end point of each mobile robot, and controlling each mobile robot to move from the movement starting point according to the environmental grid map information;

[0009] S2: randomly selecting a candidate node after the starting point for each mobile robot according to the roulette wheel method, and using the improved potential field heuristic function to determine the corresponding optimal mobile node, controlling each mobile robot to move according to the optimal mobile node until the corresponding mobile robot reaches the end point, or cannot move further, and at the same time determining whether all mobile robots have reached the end point. If any mobile robot has not reached the end point, re-execute step S2 until all mobile robots have reached the end point;

[0010] S3: Based on the candidate nodes and the optimal mobile node, obtain the moving path of the optimal mobile robot and the moving path of the worst mobile robot in step S3, as well as the moving paths of the remaining mobile robots, and record the moving path lengths of all mobile robots after the initial iteration, and then update the global pheromone concentration;

[0011] Also includes:

[0012] Obtain the pheromone concentrations of the moving paths corresponding to all mobile robots;

[0013] Updating the pheromone concentration using a pheromone concentration update formula;

[0014] The pheromone concentration update formula is:

[0015]

[0016] Where, Represents iteration The mobile robot moves from the mobile node To the candidate node The pheromone concentration, It is a dynamic global information volatile element; For the The optimal mobile robot moves from the mobile node To the candidate node pheromone concentration; For the The worst mobile robot moves from the mobile node To the candidate node pheromone concentration; For the The remaining mobile robots move from the mobile node To the candidate node pheromone concentration; For the Move the robot at iteration the length of the path taken; is the pheromone intensity, which is a constant; is the number of mobile robots that found the shortest path in the current iteration; is the number of mobile robots that found the longest path in the current iteration; Indicates the shortest path length of the current iteration; Indicates the longest path length of the current iteration;

[0017] S4: End the search and determine whether the maximum number of iterations has been reached. If so, output the optimal path. If not, return to step S2 and re-execute step S2 and subsequent steps until the maximum number of iterations has been reached.

[0018] Preferably, the improved potential field heuristic function in step S2 is:

[0019]

[0020] Where, For the current mobile node To the candidate node The Euclidean distance of Candidate nodes The inverse of the distance to the nearest obstacle, calculated as: ,in Is a candidate node distance to the nearest obstacle; is the safety threshold; Candidate nodes Connectivity evaluation, based on the number of all surrounding next-mobile nodes that can move; is the iteration weight term; The first dynamic weight coefficient, the second dynamic weight coefficient, the third dynamic weight coefficient, and the fourth dynamic weight coefficient are in sequence. Dynamic adjustment.

[0021] Preferably, the dynamic global information volatile element is updated , and dynamically adjust according to the number of iterations. The specific formula is:

[0022]

[0023] Where, is the current iteration number, is the maximum number of iterations, Represents the base of natural logarithms.

[0024] Preferably, step S3 further includes:

[0025] Determine the initial moving path based on the mobile nodes of all mobile robots, and determine all connected adjacent points on each initial moving path;

[0026] Redundant turning points are removed according to all connected adjacent points on each initial moving path, and the optimal moving path, the worst moving path, and the moving paths of the remaining mobile robots are determined according to the initial moving paths after the turning points are removed.

[0027] Preferably, it also includes:

[0028] Obtaining rectangular obstacle information in the environmental grid map information;

[0029] Intersection detection is performed based on the initial moving path and the rectangular obstacle information. When the initial moving path intersects with the rectangular obstacle information, the initial moving path is infeasible; when the initial moving path does not intersect with the rectangular obstacle information, the initial moving path is valid.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] The present invention provides a mobile robot path planning method based on an improved ant colony algorithm. First, environmental grid map information is established and the starting and ending points of each mobile robot are determined. Then, the improved ant colony algorithm is used to plan a global path for each mobile robot, and targeted secondary optimization is performed to make the final output path simpler, smoother, and shorter. This provides an effective solution for multi-robot path planning in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 This is a flow chart of a mobile robot path planning method based on an improved ant colony algorithm provided by an embodiment of the present invention;

[0034] Figure 2 Schematic diagram of path planning of the traditional ant colony algorithm provided by an embodiment of the present invention;

[0035] Figure 3 Schematic diagram of path planning of the improved ant colony algorithm provided by an embodiment of the present invention;

[0036] Figure 4 This is a comparison chart of the path planning lengths of the traditional algorithm and the improved ant colony algorithm provided by an embodiment of the present invention;

[0037] Figure 5 This is a comparison chart of the running time of the traditional algorithm and the improved ant colony algorithm provided by the embodiment of the present invention;

[0038] Figure 6 It is an iterative comparison diagram of the traditional algorithm and the improved ant colony algorithm provided by the embodiment of the present invention. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present invention are clearly and completely described in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other implementations derived by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of the present invention.

[0040] It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the conditions under which the present invention can be implemented. Therefore, they have no substantive technical significance. Any structural modification, change in proportional relationship or adjustment of size should fall within the scope of the technical content disclosed in the present invention without affecting the efficacy and purpose that can be achieved by the present invention. It should be noted that in this specification, relational terms such as first and second are only used to distinguish one entity from several other entities, and do not necessarily require or imply any actual relationship or order between these entities.

[0041] Mobile robots are increasingly used in everyday life, but traditional path planning methods are no longer sufficient for the more demanding applications of mobile robots. To address the problem of achieving optimal path planning in complex environments, this paper provides a mobile robot path planning method based on an improved ant colony algorithm. This method improves the mobile robot's path search process, resulting in a smoother, simpler, and shorter path. Furthermore, collisions with obstacles are avoided during the path planning process, ensuring that the generated wheel path is both effective and safe.

[0042] In order to more clearly introduce the above-mentioned objects, features and advantages of the present invention, further detailed description is given below with reference to the accompanying drawings and specific embodiments.

[0043] like Figure 1 As shown, an embodiment of the present invention provides a mobile robot path planning method based on an improved ant colony algorithm, comprising the following steps:

[0044] S1: Establishing environmental grid map information, and determining the movement starting point and end point of each mobile robot, and controlling each mobile robot to move from the movement starting point according to the environmental grid map information.

[0045] In this embodiment, a suitable lidar camera is selected according to the characteristics of the environment in which the mobile robot is located to scan information such as fixed obstacles, unknown obstacles, and moving obstacles in the environment in which the mobile robot is located, and at the same time realize the positioning of the mobile robot.

[0046] In this embodiment, a ROS (Robot Operating System) software environment suitable for the mobile robot is installed and configured, and a keyboard remote control is used to control the movement of the mobile robot in an indoor environment. While the mobile robot is moving, a laser radar is used to scan the environment, perform synchronous positioning and mapping, and establish environmental grid map information and the corresponding coordinate system.

[0047] In this embodiment, parameters are initialized, including the pheromone weight factor , heuristic information weight factor , pheromone volatility coefficient , maximum number of iterations , pheromone intensity etc., and determine the moving starting point of the mobile robot and end point The grid number.

[0048] S2: According to the roulette method, a candidate node after the moving starting point is randomly selected for each mobile robot, and the improved potential field heuristic function is used to determine the corresponding optimal mobile node, and each mobile robot is controlled to move according to the optimal mobile node until the corresponding mobile robot reaches the end point, or cannot continue to move, and at the same time, it is determined whether all mobile robots have reached the end point. When there is a mobile robot that has not reached the end point, step S2 is re-executed until all mobile robots have reached the end point.

[0049] In this embodiment, at time , the mobile robot is located at the mobile node , with standard probability The candidate node with the largest product of pheromone concentration and heuristic information is transferred to as the next mobile node. The mobile robot moves on the grid map established according to the environmental grid map information according to the movement rule and selects the next mobile node according to the pseudo-random proportional principle, thereby realizing the search and planning of the mobile path. The movement rule formula is as follows:

[0050]

[0051] Where, represents a candidate node; Indicates that from all mobile nodes, select The largest candidate node , is the set of all mobile nodes that can move; Indicates the pheromone weight factor Under the influence of Move the node at this moment To the candidate node The pheromone concentration, is the pheromone weight factor, which controls the influence of pheromones; represents the heuristic information weight factor Under the influence of Move the node at this moment To the candidate node The heuristic potential function, is the heuristic information weight factor, which controls the influence of heuristic information; Indicates the The probability of candidate nodes in the neighborhood at the iteration; Real-time probability.

[0052] When the real-time probability is greater than the standard probability, the next mobile node is selected according to the roulette method. The probability of each candidate node in the neighborhood at the iteration , and then the mobile robot selects the next mobile node according to the real-time probability at this time. The specific formula is as follows:

[0053]

[0054] Where, Represents the neighborhood candidate movable grid set. In this formula The larger it is, the more the mobile robot tends to choose a moving path with a high pheromone concentration; The larger it is, the more the mobile robot tends to choose a node that is close to the target point.

[0055] In this embodiment, at any time, the mobile robot moves from the current mobile node according to the above movement rules. Move to the next mobile node After that, the real-time update will be sent from the mobile node Move to the next mobile node Pheromone concentration along the path.

[0056] In this embodiment, the improved potential field heuristic function is:

[0057]

[0058] Where, For the current mobile node To the candidate node The Euclidean distance of Candidate nodes The inverse of the distance to the nearest obstacle, calculated as: ,in Is a candidate node distance to the nearest obstacle; is the safety threshold; Candidate nodes Connectivity evaluation, based on the number of all surrounding next-mobile nodes that can move; is the iteration weight term; The first dynamic weight coefficient, the second dynamic weight coefficient, the third dynamic weight coefficient, and the fourth dynamic weight coefficient are in sequence. Dynamic adjustment.

[0059] in, ; , is the current iteration number, is the maximum number of iterations; With the number of iterations change, , , , , is the preset initial weight value, Represents the initial maximum weight of the distance term, which controls the weight of the path length in the initial stage. The larger the value, the more likely the mobile robot is to choose a path with a shorter distance. represents the initial weight range of the obstacle avoidance item, Represents the initial minimum weight of the obstacle avoidance item, ensuring that obstacles are not completely ignored even in the later stages. Represents the initial maximum weight of the obstacle avoidance item, avoiding frequent collisions in the early stage; It represents the initial weight of the connectivity item, which controls the initial weight of the node connectivity evaluation. The larger the value, the more likely the mobile robot is to choose a path with more feasible nodes around it. It is the maximum value of the initial iteration weight term, which controls the balance between global exploration (in the early stage) and local optimization (in the later stage). The larger the value, the more exploration is encouraged in the early stage. is the decay rate parameter, control The exponential decay rate of The faster the descent; control The exponential growth rate of The faster it rises.

[0060] In this embodiment, in order to make the heuristic potential function more adaptable to complex environments and dynamic adjustment requirements in the ant colony algorithm, a and the maximum number of iterations The potential field inspiration function is Steer the path away from obstacles, Avoid selecting isolated nodes, improve path feasibility, and adjust the initial When, high and Emphasis on distance goals and global exploration, low and Relax obstacle avoidance requirements; dynamically iterate and adjust the late stage When low and Reduce randomness, high and Strictly avoid obstacles and optimize path quality; finally balance exploration and utilization, through Gradually shift from global exploration to local optimization to avoid falling into local optimality.

[0061] The mobile robot starts the path search and uses a loop to simulate the search process of each mobile robot. In each round of iterative search, each mobile robot is triggered from the starting point of the movement, and selects the mobile node after the starting point at any time according to the roulette method, and uses the improved heuristic function to determine the corresponding optimal mobile node. Then the mobile robot moves according to the selected optimal mobile node until the mobile robot reaches the end point or cannot move further.

[0062] S3: According to the mobile node and the optimal mobile node, obtain the moving path of the optimal mobile robot and the moving path of the worst mobile robot in step S3, as well as the moving paths of the remaining mobile robots, and record the moving path lengths of all mobile robots after the initial iteration, and then update the global pheromone concentration.

[0063] Optionally, the method further includes: obtaining pheromone concentrations of the moving paths corresponding to all mobile robots; and updating the pheromone concentrations using a pheromone concentration update formula.

[0064] In this embodiment, pheromone updates are performed on moving paths of different qualities through a pheromone reward and punishment strategy, thereby achieving better results in both global search and local optimization.

[0065] First, the corresponding mobile robots are defined as follows according to the length of the moving path traversed by the mobile robot:

[0066] Optimal Mobile Robot: For those mobile robots that have found the shortest path, the paths they have taken are considered high-quality paths, and the corresponding mobile robots are therefore called optimal mobile robots. When updating pheromones, the optimal mobile robots are given larger pheromone increments to enhance the influence of these paths. Represents iteration The optimal mobile robot from the mobile node To the candidate node The pheromone concentration is defined as:

[0067]

[0068] Where, It is a dynamic global information volatile element; Represents iteration The optimal mobile robot from the mobile node To the candidate node pheromone concentration; For the Move the robot at iteration the length of the path taken; is the pheromone intensity, which is a constant; is the number of mobile robots that found the shortest path in the current iteration; Indicates the shortest path length of the current iteration;

[0069] Worst Mobile Robot: For those mobile robots that find the longest path, the paths they travel are considered low-quality paths, and the corresponding mobile robots are called worst mobile robots. When updating pheromones, the worst mobile robots are given smaller pheromone increments to reduce the influence of these paths. Represents iteration The worst mobile robot from the mobile node To the candidate node The pheromone concentration is defined as:

[0070]

[0071] Where, is the number of mobile robots that found the longest path in the current iteration; Indicates the longest path length of the current iteration; Represents iteration The worst mobile robot from the mobile node To the candidate node pheromone concentration.

[0072] Other mobile robots: For other mobile robots, the paths they have taken are considered suboptimal. When updating pheromones, regular pheromone increments are given to other mobile robots to maintain the influence of these paths. Represents iteration other mobile robots from the mobile node To the candidate node The pheromone concentration is defined as:

[0073]

[0074] Where, Represents iteration other mobile robots from the mobile node To the candidate node pheromone concentration.

[0075] At any time, when a mobile robot moves from the current mobile node to the next mobile node, the pheromone concentration on the path will be updated in real time. After all mobile robots complete the initial iteration, the worst number of mobile robots and the optimal number of mobile robots in this iteration are recorded. At the same time, the length of the moving path of all mobile robots is recorded, and the global pheromone update is performed according to the improved global pheromone formula.

[0076] When all mobile robots complete an iterative search, the pheromone concentration of the moving path of the global optimal mobile robot is updated in real time using the improved pheromone concentration formula.

[0077] The pheromone concentration update formula is:

[0078]

[0079] In this embodiment, compared with the global information volatile element in the traditional ant colony algorithm, Is a fixed constant, which leads to unreasonable pheromone distribution when the mobile robot searches for a path. In order to avoid this defect, the global information volatile is changed from a static value to a value that is dynamically adjusted according to the number of iterations. , the specific formula is:

[0080]

[0081] Where, is the current iteration number, is the maximum number of iterations, Represents the base of natural logarithms.

[0082] By adaptively updating the global information volatiles, making them smaller at the beginning of the iteration and gradually increasing, the mobile robot is encouraged to explore more. In the later stages of the iteration, as the number of iterations increases, the global information volatiles increase, causing the pheromone concentration to increase, which enhances the guiding effect of the pheromone concentration on the mobile robot, thereby correspondingly reducing the search time of the mobile robot and better balancing the global search and local search.

[0083] Optionally, it also includes: determining an initial moving path based on the mobile nodes of all mobile robots, and determining all connected adjacent points on each initial moving path; removing redundant turning points based on all connected adjacent points on each initial moving path, and determining the moving path of the optimal mobile robot and the moving path of the worst mobile robot, as well as the moving paths of other mobile robots based on the initial moving path after the turning points are removed.

[0084] In this embodiment, the traditional ant colony algorithm is an eight-degree-of-freedom search probability algorithm derived from the foraging behavior of ants. The ant colony algorithm has a simple step size, which limits the efficiency of mobile robot path planning. To solve this problem, a secondary optimization is used to eliminate redundant turning points and reduce the path length. This application removes redundant turning points by determining the connectivity of adjacent points on the path. The specific process is as follows:

[0085] Step 1: Setting the starting point and adding nodes: Starting from the starting point of the initial moving path, take the current moving node as a node of the new path and add it to the new path;

[0086] Step 2: Finding connected adjacent points: Starting from the current mobile node, traverse backwards towards the last mobile node on the initial movement path, gradually checking whether each adjacent point in the feasible area is connected to the current mobile node. Connectivity can be understood as the absence of obstacles between two points, that is, they can be directly connected. When a connected adjacent point is found, it is set as the starting point of the new path and the next step is carried out.

[0087] Step 3: Remove redundant turning points: Once a connected adjacent point is found, use it as the new current point and continue searching for the next connected adjacent point in the direction of the starting point of the initial movement path. This will directly connect the adjacent points and eliminate redundant turning points in the initial movement path.

[0088] Step 4: Path validity verification: Before connecting adjacent points, obstacle collision detection is required. If the straight path connecting adjacent points intersects with an obstacle, it means that this path is not feasible and you need to reselect the adjacent points to connect.

[0089] Step 5: Repeat the optimization steps: Repeat steps 2 to 4 until the last mobile node of the initial moving path is reached, or until no more connected adjacent points are found.

[0090] Step 6: Output optimized path: After the above steps, the obtained path is the optimized path, in which redundant turning points have been eliminated, and the path is smoother and simpler.

[0091] Optionally, it also includes: obtaining rectangular obstacle information in the environmental grid map information; performing intersection detection based on the initial moving path and the rectangular obstacle information, when the initial moving path intersects with the rectangular obstacle information, the initial moving path is not feasible, and when the initial moving path does not intersect with the rectangular obstacle information, the initial moving path is valid.

[0092] In this embodiment, during the path optimization process, obstacle collision detection also needs to be considered. Obstacle collision detection is based on the intersection detection of straight line segments and rectangular obstacles. By detecting whether the line segments on the moving path intersect with the obstacles, it is determined whether the path is valid. If a segment in the moving path intersects with the obstacle, it means that the initial moving path is not feasible.

[0093] In this embodiment, obstacle collision detection assumes that the initial movement path and the rectangular obstacle are line segments AB and CD, respectively. The cross product of vectors is calculated to determine whether they intersect. Specifically, assume that line segment AB starts at point A and ends at point B, and that line segment CD starts at point C and ends at point D. The cross product formula is as follows:

[0094]

[0095] in, represents the vector corresponding to line segment AB, Represents the vector corresponding to line segment CD. The result of the cross product can be a scalar.

[0096] According to the result of the cross product, we can determine the intersection of two line segments:

[0097] If the cross product cross is a positive value, it means that line segments AB and CD intersect, which means that the initial moving path is infeasible.

[0098] If the cross product is zero, it means that line segments AB and CD are parallel or overlap, which means that the initial moving path is valid.

[0099] If the cross product cross is a negative value, it means that line segments AB and CD do not intersect, that is, the initial moving path is valid.

[0100] In path optimization, the cross product is used to determine whether the optimized path segment intersects with the obstacle boundary. arrive To determine whether an object intersects with the obstacle boundary, the following cross product calculation can be used:

[0101]

[0102] Where, is the maximum coordinate of the obstacle boundary. If the cross product cross is positive, it means that the path segment intersects the obstacle boundary and this path should be avoided. If the cross product cross is zero or negative, it means that the path does not intersect the obstacle and the path is feasible.

[0103] In summary, the cross product is used in path planning to determine whether line segments intersect, helping us avoid collisions between paths and obstacles, thereby ensuring that the generated path is valid and safe.

[0104] S4: End the search and determine whether the maximum number of iterations has been reached. If so, output the optimal path. If not, return to step S2 and re-execute step S2 and subsequent steps until the maximum number of iterations has been reached.

[0105] In this embodiment, when the maximum number of iterations is not reached, the taboo table in the environment grid map information is cleared, and the mobile robot is instructed to , return to step S2 and execute subsequent steps in sequence until the maximum number of iterations is reached, and output the optimal path length.

[0106] In this embodiment, Figure 2 、 Figure 3 They represent the mobile robot path planning diagram of the traditional ant colony algorithm and the mobile robot path planning diagram output by the embodiment of the present invention, where black boxes represent fixed obstacles, gray cross boxes represent unknown obstacles, and light slash boxes represent mobile obstacles. The lengths of the horizontal and vertical coordinate systems are customized according to actual conditions. Figure 2 and Figure 3 By comparison, it can be seen that after multiple iterations, the optimal path output by the improved ant colony algorithm of the present invention is smoother and simpler than that of the traditional ant colony algorithm, and the path is shorter.

[0107] In this embodiment, Figure 4The figure shows the path planning length of the original algorithm and the improved ant colony algorithm provided in this application. Experimental tests were carried out using the two algorithms, with 20 samples each. The path planning length of the original algorithm ranged from 27.5 cm to 28.2 cm, and the path planning length of the improved ant colony algorithm ranged from 25.3 cm to 26.2 cm. By comparison, it can be seen that the path planning length of the improved ant colony algorithm is reduced by an average of 2.1 cm compared to the original algorithm.

[0108] like Figure 5 The figure shows the running time of the original algorithm and the improved ant colony algorithm. 20 groups of samples were tested using the two algorithms. The time range of the original algorithm was 9.24 seconds to 10.25 seconds, and the time range of the improved ant colony algorithm was 7.25 seconds to 8.26 seconds. By comparison, it can be seen that the running time of the improved ant colony algorithm is reduced by an average of 1.9 seconds compared with the original algorithm.

[0109] like Figure 6 The figure shows the algorithm iteration training diagram of the original algorithm and the improved ant colony algorithm. Figure 6 As can be seen in the figure, the original algorithm converges slowly, reaching convergence after 78 training cycles, with an optimal length of 28.2 cm. The improved ant colony algorithm converges quickly, reaching convergence after 47 training cycles, with an optimal length of 25.5 cm.

[0110] The present invention optimizes the potential field heuristic function and updates the pheromone concentration according to the improved pheromone reward and punishment strategy, while realizing the adaptive adjustment of the global information volatiles, making the mobile robot more intelligent in the path planning process, and the obtained moving path is smoother, simpler and shorter.

[0111] The foregoing description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be readily conceived by a person skilled in the art within the technical scope disclosed herein should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A mobile robot path planning method based on an improved ant colony algorithm, characterized in that: The steps include: S1: Establishing environmental grid map information, and determining the movement starting point and end point of each mobile robot, and controlling each mobile robot to move from the movement starting point according to the environmental grid map information; S2: randomly selecting a candidate node after the starting point for each mobile robot according to the roulette wheel method, and using the improved potential field heuristic function to determine the corresponding optimal mobile node, controlling each mobile robot to move according to the optimal mobile node until the corresponding mobile robot reaches the end point, or cannot move further, and at the same time determining whether all mobile robots have reached the end point. If any mobile robot has not reached the end point, re-execute step S2 until all mobile robots have reached the end point; S3: Based on the candidate nodes and the optimal mobile node, obtain the moving path of the optimal mobile robot and the moving path of the worst mobile robot in step S3, as well as the moving paths of the remaining mobile robots, and record the moving path lengths of all mobile robots after the initial iteration, and then update the global pheromone concentration; Also includes: Obtain the pheromone concentrations of the moving paths corresponding to all mobile robots; Updating the pheromone concentration using a pheromone concentration update formula; The pheromone concentration update formula is: Where, Represents iteration The mobile robot moves from the mobile node To the candidate node The pheromone concentration, It is a dynamic global information volatile element; For the The optimal mobile robot moves from the mobile node To the candidate node pheromone concentration; For the The worst mobile robot moves from the mobile node To the candidate node pheromone concentration; For the The remaining mobile robots move from the mobile node To the candidate node pheromone concentration; For the Move the robot at iteration the length of the path taken; is the pheromone intensity, which is a constant; is the number of mobile robots that found the shortest path in the current iteration; is the number of mobile robots that found the longest path in the current iteration; Indicates the shortest path length of the current iteration; Indicates the longest path length of the current iteration; S4: End the search and determine whether the maximum number of iterations has been reached. If so, output the optimal path. If not, return to step S2 and re-execute step S2 and subsequent steps until the maximum number of iterations has been reached.

2. A mobile robot path planning method based on an improved ant colony algorithm according to claim 1, characterized in that: The improved potential field heuristic function in step S2 is: Where, For the current mobile node To the candidate node The Euclidean distance of Candidate nodes The inverse of the distance to the nearest obstacle, calculated as: ,in Is a candidate node distance to the nearest obstacle; is the safety threshold; Candidate nodes Connectivity evaluation, based on the number of all surrounding next-mobile nodes that can move; is the iteration weight term; The first dynamic weight coefficient, the second dynamic weight coefficient, the third dynamic weight coefficient, and the fourth dynamic weight coefficient are in sequence. Dynamic adjustment.

3. The mobile robot path planning method based on the improved ant colony algorithm according to claim 1, characterized in that: Update the dynamic global information volatile element , and dynamically adjust according to the number of iterations. The specific formula is: Where, is the current iteration number, is the maximum number of iterations, Represents the base of natural logarithms.

4. The mobile robot path planning method based on the improved ant colony algorithm according to claim 1, characterized in that: Step S3 further includes: Determine the initial moving path based on the mobile nodes of all mobile robots, and determine all connected adjacent points on each initial moving path; Redundant turning points are removed according to all connected adjacent points on each initial moving path, and the optimal moving path, the worst moving path, and the moving paths of the remaining mobile robots are determined according to the initial moving paths after the turning points are removed.

5. The mobile robot path planning method based on the improved ant colony algorithm according to claim 4 is characterized in that: Also includes: Obtaining rectangular obstacle information in the environmental grid map information; Intersection detection is performed based on the initial moving path and the rectangular obstacle information. When the initial moving path intersects with the rectangular obstacle information, the initial moving path is infeasible; when the initial moving path does not intersect with the rectangular obstacle information, the initial moving path is valid.

Images