A navigation system for vision-guided mobile robots

In this paper we present a vision system for an autonomous mobile robot. Our robot performs goal-reaching tasks into unknown indoor environments by using visual landmarks. Robot vision processes are performed at different levels of abstraction in order to extract useful information for robot navigation. The system, based on the HSV color space, is implemented by two groups of parallel processes: a low-level vision system performs obstacle avoidance and corridor following, while a high-level vision system extracts landmark contents for high-level planning. System performance was tested on sets of digital images supplied by the RWI B12 mobile robot camera.