Moving Target Detecting And Tracking Of Mobile Robot

Moving target detecting and tracking is one of the difficult and hot topics in mobile robot research field, and it is a key technology especially in service mobile robot’s applications. In this paper, a laser range finder is used to obtain the distance information between mobile robot and the target. The corresponding algorithms and implementation schemes are proposed respectively for moving target detecting and moving target tracking of mobile robot. The effectiveness of algorithms and the feasibility of schemes are verified by experiments. The main work of this paper is as following.(1) A laser range finder is used to obtain the distance information between the mobile robot and the target in front of the robot. The laser scan data is processed by a median and mean filter-based method. A target detecting window is designed to eliminate dates unrelated to the detected target, and the laser scanning points are classified in this window.(2) The curve-like algorithm is proposed to detect the feature of human leg. The distance between two legs when people are in a state of standing or walking can be used as the basis of detecting the target human legs. The target center need to be tracked and the distance between the mobile robot and the target center are computed based on the extracted target human legs’ data.(3) The target tracking algorithm of mobile robot based on Kalman filter is used to take the predicted position of Kalman filter as the input of robot motion control, and compute speed control variable of target tracking with mobile robot. In addition, target tracking algorithm of mobile robot based on the smoothed nearness diagram algorithm is used to compute the goal-directed and smoothly varying speed control variable.(4) The moving target detecting algorithm and tracking algorithm of mobile robot mentioned above are verified by the experiment on AS-R robot experiment platform. The results show that the relative algorithms and schemes presented in this paper are feasible and effective.