Research On Complete Coverage Path Planning Of Mobile Robots

Path planning is not only one of the most important research topics in the field of mobilerobots, but also the foundation of helping the mobile robots to realizeintelligent andindependent work. The coverage path planning of mobile robots is a kind of special form inrobot path planning, and complete coverage path planning requires the path of the robots tocover the whole area of the workspace except the obstacle. Mobile robots complete coveragepath planning is widely used in the field of industry, commerce and daily life, such as cleaningrobots, spray painting robots, geological exploration robots, mowing robots, automaticharvesting robots, windowcleaning robots and so on. Based on the deep study of the the biological stimulation neuralnetwork algorithm, a new method of algorithm of the complete coverage path planning of mobilerobot is put forward, which is combined with grid method and basic idea of neuron, and after thesimulation experiment, this algorithm has been verified feasible. In this paper, some keytechniques of mobile robot system are been studied as following:1. Various methods of environmental modeling are been researched, analyzed andcompared and their advantages and disadvantages will be discussed. Then, the modeling of theenvironment by combining with the characteristics of real life environment is built and gridmethod is used. Looking each grid as a neuron, and setting each neuron’s activity values.2. After finishing the environmental modeling, using the biological stimulation neuralnetwork algorithm to make point to point path planning and complete coverage path of themobile robots, and then mobile robots which are based on biological stimulation neuralnetwork will be improved according to the randomness of the the biological stimulation neuralnetwork for path planning. Finally, the simulation experiment will be made to verified thefeasibility of the algorithm.3.This paper proposes a path planning algorithm for mobile robots traversingthe neuronal excitation after the improvement of the biological stimulation neural networkalgorithm, and then the feasibility and validity of the algorithm is verified by simulationexperiment. Finally, this paper compares the neuron excitation and biological stimulation neuralnetwork algorithm by using the same environment mapand shows the superiority of neuronexcitation algorithm by using the performance index obtained by the comparison of simulationexperiment.