The Investigation And Development Of Freescale Smart Car And Its Debugging Platform

Smart car is an important link of Intelligent Transportation System, which contains many high-tech. Intelligent cars self-driving technology has a broad application prospects in highway management, transportation and so on. In this paper, the design of intelligent vehicle systems has been discussed under the background of "Freescale" smart car competition, including intelligent vehicle model's mechanical structure and track recognition algorithm. Also a debugging platform is studied and developed to improve the efficient of image processing algorithms and control algorithms for smart car.The main work is as follows:(1) Based on the rules of Freescale Cup smart car competition, smart car model characteristics has been studied and its mechanical structure was been designed and adjusted. Under the car theory and practice, smart car's caster angle, kingpin inclination angle and other parameters were modified in order to make the car body as a flexible implementing agency. Smart car center of gravity has also been optimized.(2) In the research of identification of track, this paper analyzes and compares the advantages and disadvantages of literature binarization algorithm and jumping binarization algorithm. Based on this a dynamic jumping algorithm has been presented that can simultaneously solve the problem of gray intensity changed gradually in the remote image and the problem of external light intensity changing.(3) A Smart Car debugging platform has been developed in this paper, which is mainly used for debugging image processing algorithms and control algorithms, in order to improve the efficiency of intelligent vehicle development. Hardware selection, making, software writing and other aspects of the platform have been described in detail in this paper. After the work of making hardware and debugging driver software is completed, the platform is used in practice.After competition, the smart car model which was made in this paper won second prize in the fourth "Freescale" Cup National Smart Car Contest. In a large number of MATLAB simulations, the dynamic jumping algorithm has better process effect. In testing, the smart car debugging development platform improved debugging development efficiency for a certain extent, reduced labor intensity of the developer.