Research on robot path planning based on fuzzy neural network and particle swarm optimization

In a certain evaluation standard, robot path planning is to find a collision-free path from the initial state to the target state in an environment with obstacles, which is one of the key research directions of intelligent mobile robots. The mathematical model of the surrounding environment is established by using the grid method. The obstacle avoidance strategy of the fuzzy neural network is proposed. The function of the obstacle avoidance is realized by searching the next feasible node by the fuzzy neural network. Aiming at the parameter optimization problem of fuzzy neural network, the improved particle swarm optimization algorithm is used to optimize the parameters of fuzzy neural network, which avoids the instability of the system caused by improper parameter selection. Simulation results verify the effectiveness of the method. The simulation results show that the path planning of mobile robot based on fuzzy neural network and particle swarm optimization achieves performance index of the minimum sum of the obstacle cost and the route cost.

1 Introduction

Robot path planning is to make a collision free path between the starting point and the target point in a given working environment [1]. The neural network has the characteristics of strong fault tolerance and adaptive learning, which can better analyze and fuse the information in the unstructured environment. Fuzzy control has the ability of logical reasoning and is more effective in dealing with structured knowledge. Combining the advantages of neural network and fuzzy control, fusing the self-learning ability of neural network and the fuzzy reasoning ability of fuzzy control, a robot path planning control algorithm based on fuzzy neural network is proposed [2]. However, the single fuzzy neural network has a relatively high precision, but it has the disadvantage of slow convergence rate, and it will fall into the local extremum at any time. So it is not suitable for the large-scale and fast search [3].