Multisensor-based Human Tracking Behaviors with Markov Chain Monte Carlo Methods

Takahiro Miyashita, Masahiro Shiomi, and Hiroshi Ishiguro

(Paper #71)


Abstract

For communication robots, it is important to find a communication partner and attract his or her attention in daily environments. In this paper, we propose a method for communication robots to detect and track a human actively in order to communicate with him or her. We apply Markov chain Monte Carlo methods (MCMC) to human detection and tracking behaviors with a humanoid robot that has four types of sensors. Thus, by utilizing our method, the robot can detect and track humans with irregular motion in complicated daily environments. While tracking a human, it tries to attract attention by verbal and nonverbal communication. We verify the validity of our method by performing experiments with a humanoid-type communication robot named Robovie.

Keywords

Humanoid Control, Application for Humanoids, Human Humanoid Interaction, Humanoid Systems

Video


[640x480, 0min21s, 3.4MB, MPEG1]