Safeguarding trafficability of a wheeled, snake-like reconnaissance robot on rough terrain by a shared control system based on fuzzy logic

Abstract

Mobile robots for inspection and surveillance of hard-to-reach and hazardous areas e.g. resulting from a building collapse in the course of a natural or man-made catastrophe have to possess enhanced rough terrain mobility capabilities. They must be able to navigate through a given environment and to avoid insurmountable obstacles. Further on, they must have the ability to negotiate a wide spectrum of obstacles including e.g. wide gaps and high steps. Finally, they must be able to traverse different forms of ground without getting immobilized by a loss of traction. These abilities can be described as a mobile robot's performance indices “maneuverability”, “terrainability” and “trafficability”. High potentials with regard to mobility in unstructured and rough environments offer snake-like, kinematically redundant robots equipped with powered wheels or tracks. These potentials are based on the systems' design. Due to their slender, modular and flexible built-up the systems are basically able to travel and maneuver through noticeable narrow passes and tunnels. Further on, their kinematic redundancy can be used for a purposeful posture and terrain adaptation to safeguard traction and the system's trafficability, respectively. Finally, the systems' modular and articulated design, both, can be used to achieve an outstanding terrainability and to be able to negotiate remarkable obstacles. The described and expected potentials have to be investigated in detail as well as evaluated in practice. The paper at hand describes a wheeled, snake-like reconnaissance robot developed for this purpose. The system is equipped with a shared control system to safeguard its efficient mobility with respect to the performance indices maneuverability, terrainability and trafficability, both, on structured and rough terrain. The developed shared control system is presented focusing on the control of the system's trafficability. The involved control strategy is based on fuzzy logic. Finally, first experimental results are presented.