A soft computing approach for collision risk assessments

In this paper, we propose a collision risk assessment system using soft computing method. By using Monte Carlo Method, we make some samples for collision risk, and we recover collision risk of whole area using neural network. Multi-layer perceptron neural network will be applied to proposed algorithm. And fuzzy system will decide the final collision risk by considering not only collision risk but also time for collision. Simulation results will show the validity of our proposed method.