AGiLe：通过可供性落地的双向潜变量规划学习鲁棒长程操作
AGiLe: Learning Robust Long-Horizon Manipulation via Affordance-Grounded Bidirectional Latent Planning
将抽象子目标落地为稠密像素级视觉可供性图来驱动动作，为视觉-动作策略提供可复用的空间定位与抗干扰机制；双向潜空间规划（后向规划器+前向评论家）可缓解长程任务中的误差累积。
原文依据 grounding abstract subgoals into dense, pixel-level visual affordances that drive action
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论长程操作任务的鲁棒执行仍然是具身智能的核心挑战，需要连贯的高层规划与可靠的底层控制。现有方法常面临两个关键局限：子目标规划中的预测误差累积，导致随时间推移的偏差叠加；以及规划-执行鸿沟，即高层抽象规划无法有效落地到连续的感知-动作空间。为解决这些挑战，我们提出了一种新颖的统一框架——可供性落地的双向潜变量规划（AGiLe）。AGiLe 引入双向潜变量规划机制，联合优化后向规划器与前向评论家。后向规划器从最终目标生成目标导向的子目标，前向评论家评估其可达性，通过长程规划中的持续一致性确保时序鲁棒性。此外，AGiLe 利用可供性作为结构引导，将抽象子目标落地为稠密、像素级的视觉可供性以驱动动作，从而弥合规划-执行鸿沟。这增强了空间鲁棒性，使系统能够有效适应语义与视觉干扰。在仿真与真实环境中的大量实证评估证实，AGiLe 显著优于强基线，相较此前最优方法取得 8.5% 的提升，并在长程操作任务中展现出强大的有效性与鲁棒性。项目网站：https://agile-long.github.io。
英文原文摘要 1611 字符
The robust execution of long-horizon manipulation tasks remains a central challenge in embodied intelligence, necessitating both coherent high-level planning and reliable low-level control. Existing approaches often encounter two critical limitations: the accumulation of prediction errors in subgoal planning, leading to compounding deviations over time; and the planning-execution gap, where high-level abstract plans fail to be effectively grounded in the continuous perception-action space. To address these challenges, we propose a novel unified framework, Affordance-Grounded Bidirectional Latent Planning (AGiLe). AGiLe introduces a bidirectional latent planning mechanism that jointly optimizes a backward planner and a forward critic. The backward planner generates goal-directed subgoals from the final objective, while the forward critic assesses their reachability, ensuring temporal robustness through sustained consistency in long-horizon planning. Furthermore, AGiLe bridges the planning-execution gap by leveraging affordance as structural guidance, grounding abstract subgoals into dense, pixel-level visual affordances that drive action. This enhances spatial robustness, enabling the system to effectively adapt to semantic and visual distractors. Extensive empirical evaluations across both simulation and real-world settings confirm that AGiLe significantly outperforms strong baselines, achieving an 8.5% improvement over prior state-of-the-art methods and demonstrating strong effectiveness and robustness in long-horizon manipulation tasks. Project website: https://agile-long.github.io.