Dynamic path tracking for visual guided vehicle based on vision information

This article presents a fast and robust method designed to detect and track a goal car from images provided by a monocular camera and a binocular camera. It establishes the environment model with the fuzzy concept and carries on the partial search in the Java platform with improvement D* algorithm. An important contribution of this work is that by using it the Visual Guided Vehicle (VGV) can complete track task fast and accurately. This fact obviates the need for off-line calibration of the eye-in-hand robotic system. We have implemented, both in simulation and in experiments, and the results are presented in this paper.