Improving Automated Driving through POMDP Planning with Human Internal States
This work addresses safety and efficiency challenges in autonomous driving for real-world applications, though it is incremental as it builds on existing POMDP methods.
The paper tackled the problem of autonomous freeway driving by using POMDP planning with human internal states to improve safety and efficiency, resulting in cutting unsafe situations by half or increasing success rates by 50% compared to MDP baselines.
This work examines the hypothesis that partially observable Markov decision process (POMDP) planning with human driver internal states can significantly improve both safety and efficiency in autonomous freeway driving. We evaluate this hypothesis in a simulated scenario where an autonomous car must safely perform three lane changes in rapid succession. Approximate POMDP solutions are obtained through the partially observable Monte Carlo planning with observation widening (POMCPOW) algorithm. This approach outperforms over-confident and conservative MDP baselines and matches or outperforms QMDP. Relative to the MDP baselines, POMCPOW typically cuts the rate of unsafe situations in half or increases the success rate by 50%.