Autonomous mobile robot navigation system designed in dynamic environment based on transferable belief model

This paper investigates the possibility of using transferable belief model (TBM) as a promising alternative for the problem of path planning of nonholonomic mobile robot equipped with ultrasonic sensors in an unknown dynamic environment, where the workspace is cluttered with static obstacles and moving obstacles. The concept of the transferable belief model is introduced and used to design a fusion of ultrasonic sensor data. A new strategy for path planning of mobile robot is proposed based on transferable belief model. The robot’s path is controlled using proposed navigation strategy that depends on navigation parameters which is modified by TBM pignistic belief value. These parameters are tuned in real time to adjust the path of the robot. A major advantage of the proposed method is that, with detection of the robot’s trapped state by ultrasonic sensor, the navigation law can determine which obstacle is dynamic or static without any previous knowledge, and then select the relevant obstacles for corresponding robot avoidance motion. Simulation is used to illustrate collision detection and path planning.

► We study a path planning strategy for robot using transferable belief model (TBM).
► Sonar data interpreted by TBM are collected and updated to build a map.
► The proposed strategy can determine which obstacle is dynamic or static.
► Anti-deadlock mechanism based on TBM is proposed in this paper.
► The proposed strategy is used to realize avoidance motion for robot successfully.