In this paper, we propose a mobile robot architecture for person tracking, consisting of an active stereo vision module (ASVM) and a navigation module (NM). The first uses a stereo head equipped with a pan-tilt mechanism to track a moving target (selected by an operator) and keep it centered in the visual field. Its output, i.e. the 3D position of the person, is fed to the NM, which drives the robot towards the target while avoiding obstacles. For this, a hybrid navigation algorithm is adopted with a reactive part that efficiently reacts to the most recent sensor data, and a deliberative part that generates a globally optimal path to a target destination, such as the person's location. As a peculiarity of the system, there is no feedback from the NM or the robot motion controller (RMC) to the ASVM. While this imparts flexibility in combining the ASVM with a wide range of robot platforms, it puts considerable strain on the ASVM. Indeed, besides the changes in the target dynamics, it has to cope with the robot motion during obstacle avoidance. These disturbances are accommodated via a suitable stochastic dynamic model for the stereo head-target system. Robust tracking is achieved by combining a color-based particle filter with a method to update the color model of the target under changing illumination conditions. The main contributions of this paper lie in (1) devising a robust color-based 3D target tracking method, (2) proposing a hybrid deliberative/reactive navigation scheme, and (3) integrating them on a wheelchair platform for the final goal of person following. Experimental results are presented for ASVM separately and in combination with a wheelchair platform-based implementation of the NM.
|Number of pages||20|
|Journal||Integrated Computer-Aided Engineering|
|Publication status||Published - 24 Jul 2006|
Bibliographical noteIntegrated Computer-Aided Engineering, Vol. ?, Nr. ?, pp. ?, .
- mobile robot
- active vision