The implementation is based on MIT-licensed highway env and Driving-IRL-NGSIM.
sanjulavj / multi-behavior-learning-for-socially-compatible-autonomous-driving Goto Github PK
View Code? Open in Web Editor NEWProposed to leverage human driving data to learn a social preference model of human driving using inverse reinforcement learning and then integrated it with reinforcement learning-based Autonomous Vehicles policy synthesis using Social Value Orientation (SVO) theory to learn diverse levels of socially compatible driving.