PROJECT TITLE :

Sensor Fusion Algorithm Design in Detecting Vehicles Using Laser Scanner and Stereo Vision

ABSTRACT:

It is well known that laser scanner has higher accuracy than stereo vision in detecting the distance and velocity of the obstacles, whereas stereo vision will distinguish the objects higher than the laser scanner. These advantages of every sensor will be maximized by sensor fusion approach thus that the obstacles in front can be detected accurately. In this paper, high-level sensor fusion for the laser scanner and stereo vision is developed for object matching between the sensors. Time synchronization, object age, and reordering algorithms are designed for strong tracking of the objects. A time-delay update algorithm is additionally developed to determine the method time delay of the laser scanner. The expanded laser scanner data at each 1 ms is predicted by Kalman filter and is matched with the stereo vision information at every sixty six ms. A cost operate is formulated to explain the article matching similarity between the sensors, and the simplest matching candidate is chosen for the minimum value function. The proposed matching algorithms are verified experimentally in field tests of varied maneuvering cases.