A MULTI-CAMERA VIEW DIRECTION PLANNING STRATEGY FOR MOBILE ROBOTS

摘要

A multi-camera view direction planning strategy for mobile robots is discussed. Two concurrent tasks for efficient and safe locomotion in dynamic environments are considered: self-localization and dynamic object tracking. The approach is to assign different tasks to different cameras, such that for each task an individual optimal view direction is selected based on information gain maximization. Thereby, the individual task performance is improved significantly. The performance of the proposed strategy is evaluated in simulations considering a humanoid robot navigation scenario and compared with two conventional multi-camera view direction planning strategies.