Camera-based indoor navigation for service robots

In our current research, we are developing a practical mobile vehicle navigation system which is capable of controlling multiple vehicles on the general environment. In order to realize the practical use, navigation system should have high accuracy and be low cost. Therefore, in this paper, we propose novel vehicle navigation system which realizes multiple vehicles navigation using high accuracy localization scheme by ceiling TV cameras with infrared filters and LED markers on target vehicles. Our key ideas are two topics. One topic is system integration which consists of vehicle localization system and CORBA network. Using our ideas, our system can extend navigation areas and target vehicles with low cost. The other topic is novel navigation scheme under occlusion condition. For practical use, mobile vehicles are required to continue its tasks with safety on temporary occlusion condition. Our developed system can adjust navigation path and velocity of target vehicle based on estimation odometry error. The performance of our proposed control scheme is experimented by experiments in general environment.