Development Of Intelligent Public Transport Safety Driving Monitor System

Traffic safety is a perpetual theme in the development of automobile traffic, with the rapid increase of vehicle population, road traffic accidents, particularly the high incidence of serious accidents; safety problems have become increasingly prominent. The passengers of bus or coach will be injured badly in the vehicle accident because there are always too many passengers on the vehicle. So it is very important for us to improve and develop vehicle especially the public traffic vehicles security guard technology.This article embarks from the vehicle active safety, has carried on some positive beneficial explorations in the vehicles safety ensure area of technology. Based on the key technology of the National Program 863"Research on the intelligent navigation system of farm machine based on information fusion of mutlisensors", some special improvement have been made according to the different research object of this article. It is not difficult to extract simple information such as road lane mark and the relative position of vehicle on the road from the machine vision sensor, so we have made some improvement on the lane mark identification and tracking technology including using a method based on Zernike Moment line detection algorithm, and then use lane mark features to extract the lane marking feature point, and use linear regression method to extract lane mark parameters. Hough transform was used to extract the lane mark parameters through mapping the line of image interspace to navigation parameters interspace.According to the image lane marking parameter information,not the position and direction of the parameter estimation model in the current traffic lane were established, but also the establishment of a lane width of the model. These models can be used to estimate the current parameters of the position, direction and lane width in current vehicle lanes, and make the Lane Departure Warning possible.The arrival time prediction method of intelligent vehicle advanced in this article is based on unary linear regression. Though this method, the effect factor of traffic condition was drew vie unary linear regression analysis, and the downstream velocity can be forecasted according to the current velocity of vehicle. The forecasting accuracy of vehicle's arriving time improved because many factors such as arrival time had been considered.In-vehicle security monitoring and control device was developed, and this device is provided with the function of Lane Departure Warning and arrival time prediction. Meanwhile, Local information display equipment with the function of displaying arrival time and the vehicle's real-time position was developed. The research goal is to develop one kind of Lane Departure Warning System basically drawing close to the practical ness and provide the theoretical and technical support for our country Automotive Safety Driving Assist System.