CtRL-Sim: Reactive and Controllable Driving Agents with Offline Reinforcement Learning
This addresses the need for more realistic and controllable simulation environments for autonomous vehicle testing, though it is an incremental improvement over existing methods.
The paper tackled the problem of generating reactive and controllable traffic agents for autonomous vehicle simulation by proposing CtRL-Sim, which uses return-conditioned offline reinforcement learning to enable fine-grained manipulation of agent behaviors, resulting in the generation of realistic safety-critical scenarios.
Evaluating autonomous vehicle stacks (AVs) in simulation typically involves replaying driving logs from real-world recorded traffic. However, agents replayed from offline data are not reactive and hard to intuitively control. Existing approaches address these challenges by proposing methods that rely on heuristics or generative models of real-world data but these approaches either lack realism or necessitate costly iterative sampling procedures to control the generated behaviours. In this work, we take an alternative approach and propose CtRL-Sim, a method that leverages return-conditioned offline reinforcement learning (RL) to efficiently generate reactive and controllable traffic agents. Specifically, we process real-world driving data through a physics-enhanced Nocturne simulator to generate a diverse offline RL dataset, annotated with various rewards. With this dataset, we train a return-conditioned multi-agent behaviour model that allows for fine-grained manipulation of agent behaviours by modifying the desired returns for the various reward components. This capability enables the generation of a wide range of driving behaviours beyond the scope of the initial dataset, including adversarial behaviours. We show that CtRL-Sim can generate realistic safety-critical scenarios while providing fine-grained control over agent behaviours.