Research map / Vision-language and multimodal / Robot action and navigation
CogDriver: Integrating Cognitive Inertia for Temporally Coherent Planning in Autonomous Driving
Pei Liu, Qingtian Ning, Xinyan Lu, Haipeng Liu, Weiliang Ma, Dangen She, Xianpeng Lang, Jun Ma
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 temporal memory module (TCM) (ablation) | CogDriver-Agent · nuScenes open-loop | L2: 0.38 -> 0.34 | Improved |
| add TCM (ablation) | CogDriver-Agent · nuScenes open-loop | collision rate: 0.44 -> 0.40 | Improved |
| add TCM (ablation) | CogDriver-Agent · nuScenes open-loop | infraction rate: 3.65 -> 3.18 | Improved |
| add TCM (ablation) | CogDriver-Agent · CogDriver-nuScenes VQA | BLEU-1: 52.24 -> 51.54 | Hurt |
| add TCM (ablation) | CogDriver-Agent · CogDriver-nuScenes VQA | recall: 54.86 -> 55.19 | Improved |
| use full set of language components (environment, dynamic, static, reasoning, action) (ablation) | CogDriver-Agent · CogDriver-nuScenes VQA | BLEU-1: 47.91 -> 51.54 | Improved |
| vary language components (ablation) | CogDriver-Agent · nuScenes open-loop | L2: 0.34 across all configs | No clear effect |
| replace baselines with CogDriver-Agent (cognitive inertia) | Bench2Drive prior best · Bench2Drive closed-loop | Driving Score: 64.22 -> 78.21 | Improved |
| CogDriver-Agent versus DriveTransformer on open-loop L2 | Bench2Drive · Bench2Drive | Avg. L2: 0.62 vs 0.63 | No clear effect |
Closest papers by meaning
- Drive-R1: Bridging Reasoning and Planning in VLMs for Autonomous Driving with Reinforcement Learning (AAAI 2026 · similarity 0.89)
- $AutoDrive\text{-}P^3$: Unified Chain of Perception–Prediction–Planning Thought via Reinforcement Fine-Tuning (ICLR 2026 · similarity 0.89)
- OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanation (NeurIPS 2026 · similarity 0.89)
- Explicit Geometric Chain-of-Thought for Vision-Language-Action in Autonomous Driving (NeurIPS 2026 · similarity 0.89)
- LaST-VLA: Thinking in Latent Spatio-Temporal Space for Vision-Language-Action in Autonomous Driving (NeurIPS 2026 · similarity 0.89)
- DriveSpatial: A Benchmark for Spatiotemporal Intelligence in VLMs for Autonomous Driving (NeurIPS 2026 · similarity 0.88)
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.
