CN-121977571-A - Multi-dimensional path planning method of mobile robot based on AROA improved optimization algorithm

CN121977571ACN 121977571 ACN121977571 ACN 121977571ACN-121977571-A

Abstract

The invention relates to a multi-dimensional path planning method of a mobile robot based on AROA improved optimization algorithm, which comprises the steps of obtaining a task scene of the mobile robot, establishing a grid map model by utilizing the task scene, marking a path starting point, a target point and barrier information, constructing a multi-objective cost function, carrying out path planning on the grid map model by adopting the improved optimization algorithm based on AROA, obtaining the path planning by utilizing the improved optimization algorithm based on AROA and utilizing the AROA optimization algorithm to fuse the BKA optimization algorithm, calculating the path fitness of population individuals by utilizing the multi-objective cost function in the path planning process, and taking the individual with the minimum path fitness as an optimal path individual to be used for obtaining the optimal path of the mobile robot. The method can effectively solve the problems that the path planning of the mobile robot in a complex multidimensional environment is easy to sink into local optimum and the convergence speed is low, and remarkably improves the accuracy and engineering practicability of the path planning.

Inventors

CHEN FENG
YANG ZEBIN
XIAO JUNHONG
GUO FUJUN
Xiao Zhongkuang
WANG ZHIHONG

Assignees

广东工业大学

Dates

Publication Date: 20260505
Application Date: 20260209

Claims (9)

1. A mobile robot multidimensional path planning method based on AROA improved optimization algorithm, comprising: Acquiring a task scene of the mobile robot, establishing a grid map model by using the task scene, and marking a path starting point, a target point and barrier information; constructing a multi-objective cost function, and planning a path of the grid map model by adopting a AROA-based improved optimization algorithm, wherein the AROA-based improved optimization algorithm is obtained by utilizing a AROA optimization algorithm to fuse a BKA optimization algorithm; In the path planning process, calculating the path fitness of the population individuals by using the multi-objective cost function, and taking the individual with the minimum path fitness as the optimal path individual for acquiring the optimal path of the mobile robot.
2. The mobile robot multidimensional path planning method based on AROA improved optimization algorithm of claim 1, wherein the obstacle information comprises threat radius, equipment size and risk extension distance.
3. The mobile robot multidimensional path planning method based on AROA improved optimization algorithm of claim 1, wherein constructing the multi-objective cost function comprises: ; Wherein, the As the weight coefficient of the light-emitting diode, For a multi-objective cost function, In order for the path length cost to be high, In order for the obstacle to avoid the cost, In the cost of the potential energy, Is a path smoothing cost.
4. A mobile robot multidimensional path planning method based on AROA improved optimization algorithm as claimed in claim 3, wherein the path length cost comprises: ; Wherein, the As a total number of path nodes, Is the first The coordinates of the nodes of the path, Is the first Coordinates of the path nodes; The obstacle avoidance costs include: ; Wherein, the As a total number of obstacles, For the shortest distance of an obstacle to a path segment, In order to threat the radius of the circle, In order to be able to size the device, A distance is extended for danger; the potential energy cost penalty includes: ; Wherein, the The total potential energy of the path node is obtained by superposing repulsive potential energy and gravitational potential energy; the path smoothing cost includes: ; Wherein, the Is a vector of adjacent path segments.
5. The method for mobile robot multidimensional path planning based on AROA's improved optimization algorithm of claim 1, wherein using AROA's improved optimization algorithm to perform path planning on the grid map model comprises: step 1, setting a population scale and a maximum iteration number, generating an initial population through a mixed chaotic mapping seed model, and screening initial optimal individuals meeting grid boundary constraint in the initial population; Step 2, carrying out population iterative updating on the initial optimal individuals, namely calculating energy factors, executing global exploration when the energy factors are larger than a target threshold, executing local development when the energy factors are not larger than the target threshold, expanding search boundaries by an improved FADs behavior simulation mechanism, and reserving historical optimal individuals by combining a memory mechanism; and step 3, repeating the step 2 until the maximum iteration times are reached, calculating the path fitness of the population individuals by using the multi-objective cost function, and taking the individual with the minimum path fitness as an optimal path individual for acquiring the optimal path of the mobile robot.
6. The mobile robot multidimensional path planning method based on AROA improved optimization algorithm of claim 5, wherein performing the global exploration includes generating a direction vector based on AROA attraction-repulsion operator and generating a search step in combination with a Lewy flight for expanding a population search range: ; l(s) is the Laiwei flight search step length, and controls the span of population global exploration; The method is characterized in that the method is a gamma function and is used for generating random step length conforming to the Lewy distribution, beta is a step length adjusting parameter, the value range is 1.0-1.8, the randomness and the stability are balanced and searched, and s is a distance factor of individuals in the current population and the history optimal individuals.
7. The method for planning a multi-dimensional path of a mobile robot based on AROA improved optimization algorithm as set forth in claim 5, wherein performing the local development includes simulating BKA attacks and migration behaviors to generate candidate locations and introducing a cauchy distribution for local fine tuning; The probability density function of the cauchy distribution is: ; Wherein, the As a parameter of the dimensions of the device, Is a location parameter.
8. The mobile robot multidimensional path planning method based on AROA improved optimization algorithm of claim 5, wherein the improved FADs behavioral simulation mechanism expands the search boundaries comprising: ; wherein J is a random matrix, the element value is 0-1, which is used for generating a random search direction, For a logical matrix element of 0 or 1, for screening for a valid search position, 1 means that the position is reserved, 0 means discard, 0-1 Random number, for switching between two search modes, For two randomly selected individuals, for introducing population diversity, Is the position coordinates of the population individuals after the (i+1) th iteration, For the population individual position coordinates at the ith iteration, For the decay factor, t is the current iteration number, iter is the maximum iteration number, for reducing late search fluctuations, Is the minimum coordinate boundary of the individual position of the population, is determined by the effective range of the grid map, And the maximum coordinate boundary of the individual positions of the population is determined by the effective range of the grid map.
9. The mobile robot multidimensional path planning method based on AROA improved optimization algorithm of claim 5, wherein iteratively updating the initial optimal individual population further comprises: Calculating AROA decay parameters for adjusting global exploration intensity: ; Calculating BKA inertial weights for balancing local development weight: ; Wherein, the For the current number of iterations, Is the maximum number of iterations.

Description

Multi-dimensional path planning method of mobile robot based on AROA improved optimization algorithm Technical Field The invention relates to the technical field of autonomous navigation of mobile robots, in particular to a mobile robot multidimensional path planning method based on AROA improved optimization algorithm. Background The mobile robot has been widely used in various fields such as industry and civil use by virtue of the characteristics of high flexibility and strong environmental adaptability, one of the core technologies of autonomous operation is path planning, namely, in the environment with obstacles, an optimal path meeting the requirements of path shortest, obstacle avoidance safety and smooth movement is designed, and the operation efficiency and task reliability of the robot are directly affected. In the prior art of path planning, conventional algorithms such asThe algorithm and Dijkstra algorithm are simple in principle, the calculation complexity is exponentially increased along with the environment complexity, the real-time performance is poor in a large-scale multi-obstacle scene, the task requirement of dynamic change is difficult to adapt, the sampling-based algorithm (such as a fast-expansion random tree algorithm) can cope with a high-dimensional environment, the path quality is dependent on subsequent optimization, redundant turning is easy to occur, and the movement energy consumption and the control difficulty of the robot are increased. In recent years, meta heuristic optimization algorithm becomes a research hot spot in the field of path planning, wherein the attraction-repulsion optimization algorithm (AROA) maintains population diversity through attraction-repulsion among individuals, has prominent global exploration capacity, can effectively cover the search space of a complex environment, but has the defects of insufficient local development precision and low search efficiency in the later convergence period, and the black-wing iris algorithm (BKA) simulates attack and migration behaviors of the black-wing iris, has excellent local optimization performance, can quickly fine-tune path details, has limited global search range, is easy to fall into local optimal solution, and causes the problem of inflexible obstacle avoidance or path redundancy of a planned path. In addition, the path cost function of the existing method focuses on single constraint in multiple ways (for example, only the path length or obstacle avoidance safety is considered), key engineering requirements such as potential energy cost and path smoothness are ignored, so that the planned path is insufficient in practicability, meanwhile, the population initialization is mostly in a random distribution mode, diversity is lacking, the searching efficiency and the optimizing precision of an algorithm are further limited, and the engineering application requirements of the mobile robot in a complex scene cannot be met. Disclosure of Invention Aiming at the defects that a single optimization algorithm in the prior art generally has low convergence speed, is easy to fall into local optimum and has insufficient planned path practicability in a mobile robot path planning scene, the invention provides a mobile robot multidimensional path planning method based on AROA improved optimization algorithm, and realizes efficient, safe and smooth path planning of a mobile robot in a complex environment. In order to achieve the above object, the present invention provides the following solutions: A mobile robot multidimensional path planning method based on AROA improved optimization algorithm, comprising: Acquiring a task scene of the mobile robot, establishing a grid map model by using the task scene, and marking a path starting point, a target point and barrier information; constructing a multi-objective cost function, and planning a path of the grid map model by adopting a AROA-based improved optimization algorithm, wherein the AROA-based improved optimization algorithm is obtained by utilizing a AROA optimization algorithm to fuse a BKA optimization algorithm; In the path planning process, calculating the path fitness of the population individuals by using the multi-objective cost function, and taking the individual with the minimum path fitness as the optimal path individual for acquiring the optimal path of the mobile robot. Optionally, the obstacle information includes threat radius, equipment size, and risk extension distance. Optionally, constructing the multi-objective cost function includes: ; Wherein, the As the weight coefficient of the light-emitting diode,For a multi-objective cost function,In order for the path length cost to be high,In order for the obstacle to avoid the cost,In the cost of the potential energy,Is a path smoothing cost. Optionally, the path length cost includes: ; Wherein, the As a total number of path nodes,Is the firstThe coordinates of the nodes of the path,Is the firstCoordinates o