The Research Of Vision-based Intelligent Vehicle Design And Practice

AGV is a kind of Intelligent Vehicle. Vision-based AGV designed by oneself adopts the vision servo control system of the image characteristic. It uses the position deviation and angle deviation of the road boundaries as input control measure, making use of the fuzzy control method, proceeds the adjustment of AGV positions continuously to force AGV following the road boundaries. Vision-based AGV control system has two subsystems. One is Road boundaries identified system and the other is driving wheel control system. The two subsystems are connected by RS-232. The road boundaries identified system solves the antinomy between the accurate and real-time processing of image identified. A method about image processing is put forward with simple principle including the improved image segmentation of optimal threshold, the little data-processing filter based on border detection. Driving wheel control system can stably receive the control instruction sending out by the road boundaries identified system, making use of the PWM method to drive motor proceed the act of accelerating, deceleration or brake according to the receiving control instruction. In order to testify the correctness of the tracking strategy and the working effect of the vision-based AGV, experiments are made in different tracking path. The experiment results show that the vision-based AGV designed by oneself has the strong capability against noise disturbance and can carry out the path tracking accurately and reliably.