Person identification by mobile robots in indoor environments

This paper addresses the problem of identifying persons with a mobile robot. In the proposed system, people are first detected and then tracked with the robot´s laser range-finder sensor, using an independent Kalman filter for each person. After segmentation, the rectangular region of the image containing the person is divided into regions corresponding to the person´s head, torso and legs. Colour features are extracted from each region for input to a pattern recognition system. Five alternative classification methods were investigated, including experiments on a real robot and with a static camera system. The best identification performance was obtained with an ensemble of self-organizing maps (ESOM), where one self-organizing map is trained for each person in the robot´s database. We also discuss how to incorporate the new method into a complete application of a robotic security guard.