Supervisory control of autonomous emergency braking with active learning for active safety

Autonomous Emergency Braking is bringing significant improvements to automotive safety by autonomously braking the vehicle before a collision occurs. Yet current commercial systems underperform, particularly in the presence of uncertainty of the vehicle state and road surface parameter estimation. Our method, Active Learning for Active Safety Autonomous Emergency Braking (ALAS-AEB), addresses these issues by developing a complete AEB system. By the virtue of modelling each component as a Discrete Event System, this paper exploits Supervisory Control Theory to design a monolithic AEB supervisor, supported by an active learning approach, to prevent all possible collisions. It is evaluated and verified by the ISO22733 test standard, where a City Scenario is selected to demonstrate the benefits of the proposed method. This work illustrates the advantages of using Active Learning in ALAS-AEB, effectively mitigating collisions where existing state-of-the-art methods fail to do so.