Abstract
In this thesis, the artificial potential field method is used to develop obstacle avoidance algorithm for autonomous wheeled robot and the discrete sliding mode control algorithm is applied to design motion controller for the robot. We designed an embedded obstacle avoidance system for autonomous mobile robot using laser range finder to scan environment range data sensor and implement the system on the microcontroller. The artificial potential field approach established a path planning method for mobile robot moving to target without collision by introducing an attractive potential field at a target and a repulsive potential field at an obstacle which is closed to robot. Here, we designed a modified attractive potential field to improve the slow moving rate problem when robot is approaching the target based on the artificial potential field method . The obstacle avoidance control system and robot motion control system were implemented separately using two different microcontrollers. We devised a protocol that transmits the Hokuyo URG04-LX laser range finder commands and decodes the range data as well. This system is combined with the obstacle avoidance algorithm to construct an obstacle avoidance strategy determination system. The system can also process the range data when the laser range finder is transferring scan data to our system and can manage a set of scan of range data and then use this data to compute the obstacle avoidance strategy and provide motion command to robot within a period of 0.1 second. The robot platform of this thesis is equipped with two brushed DC motors as actuators. To claim robustness of mobile robot control, this thesis applied discrete-time sliding mode control scheme to both DC motors. The robot motion control command system and robot coordinate and orientation calculation method based on trapezoid velocity profile are designed. The experiments carried out are straight-line movement of robot, the simulation and implementation of modified artificial potential field method and obstacle avoidance in indoor environment, and the results further verified the performance.