A novel method for providing situational awareness to longitudinal controllers in (semi-) automated vehicles

This work examines how a classifier's output of a long-term, motivation-based cut-in prediction can be mapped to a (semi-) automated car's reaction to it. A novel approach to decision making is proposed automated longitudinal guidance systems based on situational contexts. The decider has been implemented in BMW's current automated driving framework and we show that the algorithm exhibits more comfortable and cooperative behavior to cut-in vehicles and, when combined with existing short-term deciders, that it slightly improves the short-term prediction times as well.