Man-machine Integration Automotive Safety Warning Research Based On The Obstacles Intelligent Identification

With the rapid growth of the national economy, more and more people have owned cars, which bring great convenience to us for travel. However, traffic accidents occur in high speed growth too, and the loss make us aching. So the technology of vehicle active safety has attracted more and more attention. Intelligent Transportation System (ITS) has become a world topic for discussion. Our country began followed international ITS technology since 1990's. But there is still a gap between current technology and actual application. So governments and research institutions around the world are developing this research, and some key techniques of it have been studied in this paper.Intelligent recognition for obstacles in front of the vehicle is first researched, and introduces a new method for target classification and identification using Radar Cross Section (RCS) of targets. There have been numerous reports about target identification only to the target state. However, it is important to anti-collision system if we can identify the specific classes and shapes of obstacles. It provides basic information to system to decide "How should do next?". Detailed analysis has been done in this paper, including the definition of RCS, calculate methods and the target character information it covers. In addition, the classification and identification method this paper used (fuzzy control) is introduced. The feasibility of the new method proposed this paper is proved by simulation.Working principle of Frequency modulated continuous wave (FMCW) has been investigated to detect the possible danger target state. The method of distance and velocity measurement of radar systems was specifically analyzed. There are many reports on this filed, so we borrow from their research results to calculate the distance and the velocity between our vehicle and obstacle.The ultimate goal of detecting obstacles information is to build a forewarning system. And the system should incorporate driving features. Driving behavior characteristic has been analyzed detailedly in this paper, and a People - Machine - Environment model been established. Then the safe distance between vehicle and obstacle can be calculated with the information gotten, which shows the characteristics of Man-machine integration. The safe distances under various conditions (weather conditions, road conditions, different speed and the relative speed) have been simulated using Matlab language. The influences of various factors on the safe distances were analyzed. The results show that the method brought forward in this paper is feasible.