Research On Path Planning Method Based On Improved Sparrow Search Algorithm Combined With Potential Field Metho

Path planning is an important topic in the field of artificial intelligence,which is a core technology for enabling autonomous decision-making and control of intelligent agents.Path planning methods can be classified into global path planning algorithms and local path planning algorithms.Global path planning algorithms are unable to avoid temporary dynamic or static obstacles,while local path planning algorithms have long planning times,low efficiency,and lack real-time capabilities.This article aims to propose a novel hybrid path planning algorithm to address the limitations of both global and local path planning algorithms and assist intelligent agents in finding optimal or suboptimal paths from a starting point to a destination.The research content of the paper is as follows:(1)Intelligent agent global path planning: To address the shortcomings of the traditional sparrow search algorithm,such as insufficient optimization capability,weak local search capability,a tendency to get trapped in local optima,and lack of stability,the article proposes an improved sparrow search algorithm based on a hybrid strategy.First,an improved Circle chaotic mapping is introduced to initialize the population,increasing population diversity and improving the algorithm’s optimization capability.Then,the position updating method for the algorithm’s explorers is improved by introducing an adaptive periodic convergence factor ,enhancing global search capability and improving convergence speed.The position updating method for the algorithm’s followers is also improved to increase local search capability and improve convergence speed.The position updating method for the algorithm’s sentinels is improved to prevent the algorithm from getting trapped in local optima.Finally,a linear decay mechanism is introduced to improve the algorithm’s stability and local exploration capability.(2)Intelligent agent local path planning: To address the problems of target inaccessibility and local minima in traditional artificial potential field methods,the article proposes a series of improvement strategies based on the causes of these problems.To solve the issue of target inaccessibility,the article suggests adding an attractive coefficient to the repulsive potential field function,altering the force balance of the intelligent agent and helping it reach the target smoothly.To avoid getting trapped in local minima,the article proposes methods such as adding virtual target points,adding obstacle influence factors,canceling obstacle repulsion forces,and using concave obstacle completion.Adding virtual target points can help the intelligent agent break out of local equilibrium states,while adding obstacle influence factors can help the agent find better paths in complex environments.Canceling obstacle repulsion forces can prevent the agent from being trapped in narrow areas,and using concave obstacle completion can help the agent bypass concave obstacles.(3)Intelligent agent hybrid path planning algorithm: Despite the improvements made by the hybrid strategy-based sparrow search algorithm in global path planning,there are still shortcomings such as excessively large turning angles and insufficient smoothness in path curvature.To address these issues,the article proposes an adaptive path smoothing algorithm that performs curve fitting and smoothing on the original path,reducing the turning angles at each turning point.To overcome the limitation of global path planning algorithms in handling temporary obstacles in real-time,the article proposes a hybrid path planning algorithm that assists intelligent agents in dynamically avoiding temporary obstacles.