Abstract:
While there has been significant progress in deploying autonomous vehicles (AVs) in urban driving settings, there remains a long-tail of challenging motion planning scenarios that must be addressed before truly driverless operation is possible. The current paradigm for motion planner design is engineering intensive, making it challenging to scale to address the long-tail. In this work, we explore the use of offline model-based reinforcement learning as an alternative approach for designing AV motion planners. We propose two offline RL algorithms, which make use of the structure in the AV domain to create dynamics models and policies that generalize successfully. We demonstrate that these results achieve state-of-the-art performance on self-driving benchmarks and propose a vision for how these could be incorporated into a real-world AV stack.

Thesis Committee:
Jeff Schneider (advisor)
Deva Ramanan