推論から軌道幾何条件を学習する自動運転VLA
AutoVLA-D: Learn a Trajectory-Geometric Condition from Reasoning for Autonomous Driving VLA
自動運転VLA計画において、離散的な軌道幾何条件(TrajCond)を推論から学習し、連続軌道予測の精度を向上させる手法を提案した論文。
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: Yuguang Yang, Zhewen Tan, Canyu Chen, Linlin Yang, Cheng Chi, Chunyang Liu, Kehua Sheng, Jinyu Yang, Baochang Zhan, Bo Zhang, Yan Wang, Xianbin Cao
分類: cs.RO, cs.CV
原文アブストラクト
Autoregressive VLA planning usually predicts continuous trajectories as the final output, but driving instructions often describe actions rather than ego states, which weakens direct conditioning on the final continuous trajectory. We address this gap with TrajCond, which introduces a discrete trajectory-geometric condition for VLA planning. Specifically, we represent TrajCond as a 1D-Brohan differential action trajectory and learn it with a two-stage training framework before final continuous trajectory prediction. With the simultaneous autoregressive generation of discrete and numeric trajectories, we name our method AutoVLA-D (Double). Experiments on NAVSIM navtest, nuScenes, and NeuroNCAP show that discrete trajectories are weaker as direct outputs but more effective as TrajCond, consistently improving final continuous trajectory prediction across closed-loop and open-loop driving benchmarks. Code is available at https://github.com/ShivaTeam/AutoVLA-D.