Map Building Method Based On Improved D-S Evidence Theory For Mobile Robot

Focusing on the uncertainty of the environment, a grid map building method based on the multi ultrasonic sensors is provided. Firstly, the single ultrasonic sensor grid model is constructed by dividing the detected area into three regions. Then the evidences which are needed to fuse are self-adaptively approximate in order to remove the counter-intuitive behaviors. Finally, the unknown environment map is built by using the D-S evidence theory to fuse the approximate evidence. Aiming at proving the method is right, the simulation is carried out in Matlab. The simulation result shows that the method is valid in unknown environment map building.