2606.16558

Total: 1

#1 ROSA-RL: Uncertainty-Aware Roundabout Optimized Speed Advisory with Reinforcement Learning [PDF] [Copy] [Kimi2] [REL]

Authors: Anna-Lena Schlamp, Jeremias Gerner, Klaus Bogenberger, Werner Huber, Stefanie Schmidtner

Roundabouts challenge automated driving in mixed traffic, as heterogeneous and non-deterministic human behavior, unknown driving intentions, and high interaction complexity create uncertainty about whether the conflict zone will be blocked or available at the moment of entry. We present ROSA-RL -- uncertainty-aware Roundabout Optimized Speed Advisory with Reinforcement Learning. It enables safe and efficient roundabout entry for automated and human-driven vehicles in mixed traffic through probabilistic conflict forecasting. A Transformer-based model predicts conflict zone occupancy over a five-second horizon, capturing multi-agent interactions to anticipate upcoming conflicts and available gaps. The prediction outputs encode uncertainty in future motion and intent, and augment the state of a classical RL framework, enabling uncertainty-aware speed coordination. Evaluated in simulations grounded in real-world data, ROSA-RL can effectively handle uncertainty and outperform a comparable model-based baseline, closing the gap to an ideal setting assuming fully known occupancy while improving traffic efficiency and safety. The source code of this work is available under: github.com/urbanAIthi/ROSA-RL.

Subjects: Artificial Intelligence , Robotics , Systems and Control

Publish: 2026-06-15 11:00:08 UTC