Accelerating the Evolution of Personalized Automated Lane Change Through Lesson Learning

Personalization is crucial for the widespread adoption of advanced driver assistance systems. To match up with each user’s preference, the online evolution capability is a must. However, conventional evolution methods learn from naturalistic driving data, which requires a lot of computing power and cannot be applied online. To address this challenge, this paper proposes a lesson learning approach: learning from the user’s takeover interventions. By learning a lesson from real-time takeover behavior, a driving zone is generated to ensure perceived safety. Within the driving zone, a personalized trajectory is planned based on model predictive control, with an objective learned from the user’s takeover. The proposed lesson learning framework is highlighted for its faster evolution capability, adeptness at experience accumulation, assurance of perceived safety, and computational efficiency. Simulation results demonstrate that the proposed system consistently achieves successful customization without further takeover interventions. Accumulated experience yields a 24% enhancement in evolution efficiency. The average number of learning iterations is only 13.8. The average computation time is 0.08 seconds.