Dynamic Collision Avoidance Path Planning for Mobile Robot Based on Multi-sensor Data Fusion by Support Vector Machine

Statistical learning theory is introduced to movement planning of intelligent robot, considering the issues that the dynamic collision avoidance planning of mobile robot is a complicated and nonlinear system, and combine the advantages of the support vector machine (SVM) possessed, a method of mobile robot dynamic collision avoidance planning based on multi-sensor data fusion by SVM is presented in this paper. We utilize 5 ultrasonic sensors and an image sensor get environmental information in this method, and the SVM is used to do multi-sensor data fusion to compute these information, in order to achieve the purpose that dynamic control the mobile robot´s next action. The method fully utilizes the potential of the SVM and the multi-sensor data fusion to solve dynamic path planning problem of mobile robot. The simulation result shows that this method is feasible and effective.