A multimodal face detection system for elderly companion robot

Regarding to global problem of ageing, the need for robots to take care of elderly people at private homes is essential, requiring a very natural and human oriented way of interaction between the old people and robot. In this paper a multimodal face detection system is presented, which is composed of face detection based on AdaBoost learning algorithm and environmental perception method. In the experiments, the system has been successfully implemented and tested on a service robot for human face detection. It is useful for the application of human-robot interaction for elderly companion robot.