Research on vehicle's collision avoidance warning system based on trajectory data mining

In order to overcome the problems of large error of safety distance judgement and poor braking stability in traditional vehicle anti-collision warning system, this paper proposes a vehicle anti-collision warning system based on trajectory data mining. The hardware design of the system includes vehicle anti-collision warning controller, vehicle trajectory data mining sensor and vehicle anti-collision pre-warning program, which can accurately collect vehicle trajectory data. Based on the above data acquisition results, the vehicle trajectory is deeply mined and cluster analysed, and the mathematical model of critical safe distance of vehicle is constructed. The calculation principle of vehicle collision probability is used to calculate the vehicle collision probability and realise vehicle collision warning. The experimental results show that the braking stability is close to 1, which can ensure the safety of the vehicle.