Vision-based trajectory tracking algorithm with obstacle avoidance for a wheeled mobile robot

Abstract

Wheeled mobile robots are becoming increasingly important in industry as means of transportation, inspection, and operation because of their efficiency and flexibility. The design of efficient algorithms for autonomous or quasi-autonomous mobile robots navigation in dynamic environments is a challenging problem that has been the focus of many researchers dining the past few decades. Computer vision, maybe, is not the most successful sensing modality used in mobile robotics until now (sonar and infra-red sensors for example being preferred), but it is the sensor which is able to give the information ’’what” and ’’where” most completely for the objects a robot is likely to encounter. In this thesis, we deal with using vision system to navigate the mobile robot to track a reference trajectory and using a sensor-based obstacle avoidance method to pass by the objects located on the trajectory. A tracking control algorithm is also described in this thesis. Finally, The experimental results are presented to verify the tracking and control algorithms.