Highway Situation Analysis with Scenario Classification and Neural Network based Risk Estimation for Autonomous Vehicles

The development of autonomous vehicles is one of the most active research areas in the automotive industry. The objective of this study is to present a concept for analysing the situation the vehicle is currently in, providing the necessary information to high level decision making algorithms. Our work focuses on a neural network based approach for assessing risks for each of the dynamically available manoeuvres in real-time and recognizing certain traffic scenarios to prepare the decision making and affect the outcome of the behaviour planner function regarding road safety. This paper provides insight to our research we carried out in collaboration with the Commercial Vehicle Systems Division of Knorr-Bremse Fékrendszerek Kft. Given the profile of the group, the aim of our study concentrated on heavy duty commercial vehicles, although the discussed algorithms are not truck-specific.