Research map / Reinforcement learning / Visual and representation RL
Reliable Policy Transfer for Safety-Aware End-to-End Driving with Deep Reinforcement Learning
Uddin Md. Borhan, Arif Raza, Zhiliang Lin, Lu Wang, Jianqiang Li, Jie Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add uncertainty-weighted relational attention (ablation) | full framework · CARLA Town10HD | cross-track error: 0.76 -> 0.65 | Improved |
| add critic ensemble for epistemic uncertainty (ablation) | full framework · CARLA Town10HD | collisions per km: 0.009 -> 0.006 | Improved |
| add uncertainty-gated entropy (ablation) | full framework · CARLA Town10HD | collisions per km: 0.008 -> 0.006 | Improved |
| use differentiable multi-objective reward instead of event-driven penalties (ablation) | full framework · CARLA Town02 | driving score: 195.9 -> 214.3 | Improved |
| add causal-uncertainty transfer objective (ablation) | full framework · CARLA Town02 | driving score: 194.1 -> 214.3 | Improved |
| add MAML meta-initialization (ablation) | full framework · CARLA Town02 | driving score: 200.8 -> 214.3 | Improved |
| remove transfer objective or MAML (ablation) | full framework · CARLA Town10HD | cross-track error: 0.65 -> 0.65 | No clear effect |
| shift reward weights toward progress (wp 0.5) (ablation) | reward shaping · CARLA | route completion / collisions: +2.1% completion, +0.002/km collisions | Mixed |
| vary entropy base beta0 between 0.5 and 1.0 (ablation) | entropy gating · CARLA | collision rate / CTE: collisions <=0.001/km, CTE 0.03 | No clear effect |
| use ego-relational state instead of global tensor encoders | E2E driving policy · CARLA Town10HD | off-road (%): 10.8 -> 4.1 vs ST-P3 | Improved |
| transfer with aligned actions, attention and uncertainty versus target-only training | Town02 · CARLA Town02 | success rate (%): 72.1 -> 85.0 | Improved |
Closest papers by meaning
- Distill to Think, Foresee to Act: Cognitive-Physical Reinforcement Learning for Autonomous Driving (NeurIPS 2026 · similarity 0.87)
- RAD: Training an End-to-End Driving Policy via Large-Scale 3DGS-based Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- HCRMP: An LLM-Hinted Contextual Reinforcement Learning Framework for Autonomous Driving (NeurIPS 2025 · similarity 0.87)
- Model-Based Policy Adaptation for Closed-Loop End-to-end Autonomous Driving (NeurIPS 2025 · similarity 0.87)
- Uncertainty-Guided Never-Ending Learning to Drive (CVPR 2024 · similarity 0.87)
- Drive-R1: Bridging Reasoning and Planning in VLMs for Autonomous Driving with Reinforcement Learning (AAAI 2026 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
