Proposing a Sensor Fusion Technique Utilizing Depth and Ranging Sensors for Combined Human Following and Indoor Robot Navigation

One of the main challenges faced in developing autonomous navigation for indoor robots is designing the strategy to account for a combination of obstacle avoidance while fulfilling additional maneuvering requirements. An existing child-following companion robot prototype called CARMI is used as a case study for this paper. CARMI's main function is to autonomously follow a child, maintaining her within the optimal detection zone of its vision-based activity tracking system, all the while navigating through furniture and obstructions. This paper examines CARMI's requirements and surveys the state-of-the-art of indoor navigation research, culminating in a design and proposal of a sensor-fusion technique combining the depth sensor, ranging sensor array and Active IR marker tracking system. The aim is to realize a robot navigation system for both person-following and obstacle avoidance using this novel method inspired by the Wandering Standpoint Algorithm and Potential Field Method.