Crawler Type Mobile Robot Control And Obstacle Avoidance

With the rapid development of computer, electronic, mechanical, automation and artificial intelligence etc, robot research has entered a new stage. As a kind of robot, tracked mobile robot has the advantages of large support area, good performance of cross-country, good traction adhesion properties, self-resetting, strong capability of obstacle-climbing and so on. So it is suitable to work in execrable environment. More and more attentions have been paid on the study of the tracked mobile robot at domestic and international in recent years. Taking tracked mobile robot as the study subject in this paper, remote control and autonomous obstacle avoidance are achieved. Meanwhile, a new algorithm has been proposed.The main task of remote control is that the robot can monitor the surroundings and the own state of robot through the upper-Computer and wireless control the movement of the robot. In this paper, the control system is divided into microcontrollers, motor control module, communication module, sensor module and human-machine interaction module five parts. Choosing LPC2378as the main chip, the detailed design of the hardware circuit and software program have been finished followed by modularization, and then the remote control of the robot has been achieved.The primary task of autonomous obstacle avoidance is that the robot can detect the position of the obstacle and independently avoid it through the obstacle avoidance system in the case of unmanned. In this paper, the obstacle avoidance subsystem is divided into the obstacle information collection, the obstacle information processing and obstacle avoidance decision-making three major steps. Robot selected Ultrasonic Module HC-SR04as the ranging sensor, fuzzy logic method as the obstacle algorithm to fulfill the autonomous obstacle avoidance.The main objective of the new algorithm is to improve the robot’s accuracy of trajectory tracking.The values from sensors affected by noise and the state estimation of robot has error. On the basis of traditional nonlinear filtering theory, in this paper the state estimate value of robot closer to the actual value using the method of combination forward filter and backward smooth. Simulation results show that the improved algorithm greatly advanced the tracking accuracy of the robot.