Human Target Tracking using a 3D Laser Range Finder based on SJPDAF by Filtering the Laser Scanned Point Clouds

Various researches on mobile robot have been conducted and this has led to target tracking being a part of the mainstream based on mobile robot. Target tracking is the process of target recognition and data association. Target recognition is to make the feature data from a target, and it is coming from the analysis of the target data. Raw data from sensors make the target data and it could be the first recognition step for the target. Data association is about the way to track the target. It is the continuous comparison and association between previous and present target data with time. In this paper, we focus on improving target recognition performance by filtering data received from laser sensors. The key idea to the filtering process is extracting data consisting of hips from three-dimensional data. This process provides a clearer standard of distinction between people. The tracking people algorithm is based on SJPDAF to manage the occlusion situation. Experimental results from our study show that the new proposed tracking method is robust in practical environments.