Abstract:In the case that reliable plume flow direction/velocity information cannot be obtained in an indoor diffusion environment, to solve the problem of low source positioning efficiency, low success rate of the source seeking robot, a plume tracking method for source seeking robot based on gray wolf optimization algorithm is proposed. In this method, the gas concentration value is used as an individual fitness. Without the plume flow rate/flow direction sensor, the social mechanism and hunting behavior of the gray wolf population are simulated by the source-seeking robot to update the position, so that the source-seeking robot can efficiently track plume and locate source location. The gray wolf optimization algorithm, particle swarm optimization algorithm, genetic algorithm, and zigzag search strategy were used in four sets of robot plume tracking simulation experiments, and the positioning success rates of the source- seeking robot based on the gray wolf optimization algorithm were 92%, 94%, 94%, and 94%. The on-site test results show that the positioning success rates of the source- seeking robot based on the gray wolf optimization algorithm were 95%, 90%, 90%, which verifies the feasibility and effectiveness of the robot plume tracking method based on the grey wolf optimization algorithm.