Research On Obstacle Avoidance Algorithm Of Indoor Wheeled Robot

Autonomous movement is an important manifestation of intelligence and autonomy for indoor mobile robots.As the basis of autonomous movement,obstacle avoidance and path planning are unavoidable problems.It has a broad application prospect to study it.This thesis studies several key issues involved in the path planning process of indoor wheeled robots in an unknown indoor environment.The main research contents include:(1)The indoor wheeled robot system is designed,including the sensor system,the control system and the motion system.In this thesis,the method of sensor data acquisition and robot kinematics models are presented respectively,and implemented by ROS operating system.(2)Research on robot map construction,including grid map construction,obstacle detection,classification and type recognition in the environment;(3)There is a large randomness in global path planning for Rapidly-exploration Random Tree.Through regional quantization and direction guidance,the randomness of the algorithm is effectively reduced.(4)For the problem that the artificial potential field method is easy to fall into the local minimum in the local path planning.An obstacle is neglected by evaluating the influence of obstacles on the robot,the robot to effectively escape the local minimum;(5)Combining the global and local path planning algorithms to propose a hybrid algorithm to obtain an effective path in an indoor unknown situation.In the simulation and real environment,the hybrid path planning method proposed in this thesis is verified.The results show that the hybrid path planning algorithm can effectively planning path in simulation and indoor unknown environment.