Updating confidence indicators in a multi-sensor pedestrian tracking system

We present in this paper a method for confidence computing and updating in a multi-sensor pedestrian tracking system. Also this paper shows how we can profit from the data provided by four layer laser scanner to calculate detection and recognition indicators confidence on detected objects. We propose the use of Transferable Belief Model framework in order to model and combine the sensor module outputs at different times. Since detection and recognition processes are not really independent, the cautious rule is used to combine the belief functions. We propose a track confidence updating algorithm, and its interesting behavior is shown with simulated and real data acquired by four layer laser scanner.