A cooperative exploration strategy with efficient backtracking for mobile robots

Abstract

This paper proposes a cooperative robot exploration (CRE) strategy which is based on the sensor-based random tree (SRT) method. The proposed CRE strategy is for exploring unknown environments with a team of mobile robots equipped with range finder sensors. An existing backtracking technique for frontier-based exploration involves moving back through inefficient routes. To enhance the efficiency of the backtracking algorithm, a hub node is defined and the most direct backtracking route is generated using its frontier data. Numerical simulations demonstrate that the proposed strategy enables exploration of unknown environments by robots more efficiently than other common methods.