A New Hybrid Method for Mobile Robot Dynamic Local Path Planning in Unknown Environment

In this paper, a hybrid approach for efficiently planning smooth local paths for mobile robot in an unknown environment is presented. The single robot is treated as a multi-agent system, and the corresponding architecture with cooperative control is constructed. And then a new method of information fusion namely DSmT (Dezert-Smarandache Theory) which is an extension of the DST (Dempster-Shafer Theory) is introduced to deal with the error laser readings. In order to make A* algorithm suitable for local path planning, safety guard district search method and optimizing approach for searched paths are proposed. Also, the parameters of internal Proportional-Integral-Derivative (PID) controller in the goto agent are adjusted through practical experiments for the use of smoothing the path searched by optimized A* algorithm. Finally, two kinds of experiments are carried out with Pioneer 2-DXe mobile robot: one uses the hybrid method proposed in this paper, the other uses artificial potential field (APF) which is the classical algorithm for local path planning. The experimental results reveal the validity and superiority of the hybrid method for dynamic local path planning. The approach presented in this paper provides an academic support for path planning in dynamic environment with moving objects in the field.