Mr. Amit Chaulwar joined the newly established BayWISS Verbundkolleg Mobility and Transport as one of the first PhD students in 2016. After 5 years, he defended his dissertation on "Hybrid Machine Learning Methods for Vehicle Safety Applications". The PhD at THI was done in cooperation with TUM, he was supervised by Prof. Dr.-Ing. Michael Botsch (THI) and Prof. Dr.-Ing. Wolfgang Utschick (TUM).
His work proposes hybrid machine learning methods for safe trajectory planning in critical traffic scenarios. Two new model-based algorithms, namely Augmented CL-RRT and Augmented CL-RRT+, are developed by extending the sampling-based Rapid-exploring Random Tree (RRT) algorithm and combined with machine learning techniques. These algorithms plan safe trajectories considering vehicle driving dynamics characteristics while conserving on-board limited computational resources.
Trajectory planning is a key task for autonomous mobility applications. A particular challenge is the safe trajectory planning for vehicles in critical traffic scenarios with multiple static and dynamic objects. In a critical traffic scenario, a safe trajectory should be planned to avoid a collision and if this is no longer possible, to mitigate the consequences.
We congratulate him warmly and wish him much success for his future career!