Real-time 3D-tracking using acoustics-vision data fusion

The study of perception, an essential junction for autonomous robotics, has lead to new developments in the fields of modelization, data fusion and control. This paper describes a real-size real-time heterogeneous data fusion algorithm implementation on a surveillance robot, which has become necessary to validate these emerging techniques in a realistic environment. The two heterogeneous sensors concerned here are a black and white camera and an acoustic array.<
>