The Reseach On The Tracking And Collision Avoidance Control System Of The Smart Cars Based On Image Recognition

Freeway have great changes in our daily lives and national economic development with fast, economical, comfortable characteristics. But traffic accidents occured frequently. Current research on the smart cars is mainly theoretical and simulation in domestic, rarely used in the actual system. The smart model cars as a platform of the control system, which run in the prescribed path and collision avoidance algorithms on the following methods were studied, has theoretical value and practical significance.The tracking and collision avoidance system of smart cars can identify the path and targets by the monocular visual,detect outline by image processing and at last complete the target identification and distance measuremeng.Transfering information between each other via the ZigBee wireless network to complete the functions of movement in the same direction and anti-collision based on crosspriority.The image data was transmitted to the serial port on the PC by wireless, simulated by the MATLAB image processing toolbox , and the research results were used in the real smart model car.The gray image which was produced by MATLAB was extracted in equidistant method,could reduce the amount of image acquisition and keep the minimum feature.Researched on the tradition operators of edge detection and proposed a simplified edge detection operator based on threshold which can improve computing speed and extract the features of target identification and ranging . Simulation results , which was showed simple and effective , is not only simplified the tradition image processing algorithm but also substantially increase the computing speed.The information transfering among smart cars by ZigBee wireless network.ZigBee is low-power,low cost,fast speed , cost-effective and good coverage. The network has a great advantage while the data carrying capacity of the system is limited , and low hardware is required, easily applied to embedded systems.The main control chip of the tracking and collision avoidance control system in the smart cars is a microcontroller for reducing the cost . Simplification of complex image processing applications through the microprocessor is a very good attempt . In the end , the author tests the algorithms on the intelligent model car.