バックボーン内計画:運転VLMによるネイティブ連続軌道生成のためのDiffAdapterVLA
Planning in the Backbone: DiffAdapterVLA for Native Continuous Trajectory Generation with Driving VLMs
運転用VLMの後段層に軌道トークンを注入し、軽量なDiffAdapterで再帰的に軌道を洗練させることで、追加のプランナーなしに連続軌道計画を実現した手法。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Changxin Lu, Xiaoliang Meng, Yu Wu, Rui Huang, Honglin Li, Tao Chen, Kaixuan Zhou, Yadong Shao
分類: cs.RO, cs.AI
原文アブストラクト
Pretrained driving vision-language models (VLMs) integrate visual, route, language, and driving context into rich driving priors, yet their representation objectives remain separated from continuous driving planning. Existing methods typically begin trajectory generation only after the VLM has formed a final condition, leaving depth-wise condition computation outside the stepwise formation of trajectory state. We introduce DiffAdapterVLA, which realizes Planning in the Backbone: it injects explicit trajectory tokens into selected VLM late layers, bringing trajectory state into backbone forward computation, where it co-evolves with driving conditions at different depths. Lightweight layer-wise DiffAdapters organize this computation into recursive trajectory refinement, while asymmetric joint attention preserves directed guidance from the condition stream to trajectory planning. By placing planning within existing backbone computation rather than relying on an independent trajectory planner, DiffAdapterVLA adapts only lightweight trajectory modules to turn existing driving priors into efficient continuous planning capability. NAVSIM results show that it achieves high-quality closed-loop planning with low end-to-end latency using few trainable parameters, and demonstrate that jointly evolving trajectory state and depth-wise driving conditions in VLM late-layer computation effectively realizes continuous trajectory planning.