Design Method of Rough Terrain Detection and Avoidance in Unknown Environment for Space Rover

This paper describes the guidance and control method for an exploration rover in an unknown planet where a rough area exists. A mission of a space rover is sometimes interrupted when the rover falls into the rough area. To overcome this problem, we propose the method that detects and avoids rough terrain by an observer without the use of images of a camera mounted on the rover. The observer, which is based on Disturbance Accommodating Control (DAC) method, estimates nonlinear terms including running resistance and slip ratio because nonlinear term changes by rough terrain such as high soil resistance area. Therefore, the proposed method enables to detect and avoid the rough terrain automatically without the use of vision sensors. Space rover also should identify its position during a mission because of a non-GPS environment. Expanded Kalman Filter simultaneous localization and mapping (EKF-SLAM) is applied to a space rover. The numerical simulation is conducted to verify that a space rover detects and avoids the rough terrain smoothly by using the proposed method.