Don't Drop the BATON: Long-Horizon Robot Manipulation via Agentic Subtask Exploration and Transition-aware Memory
作者: Bingxin Xu, Yuzhang Shang, Emilio Ferrara
分类: cs.RO, cs.AI, cs.CV
发布日期: 2026-08-17
💡 一句话要点
提出BATON以解决长时间机器人操作中的任务链问题
🎯 匹配领域: 支柱一：机器人控制 (Robot Control) 支柱九：具身大模型 (Embodied Foundation Models)
关键词: 长时间操作 机器人操控 视觉-语言-动作 过渡感知记忆 多阶段任务 任务成功率 探索策略
📋 核心要点
- 现有方法在长时间机器人操作中面临的主要挑战是错误的累积和子任务之间的约束，导致整体任务失败。
- BATON通过将子任务作为探索单元，采用短时间范围内的探索策略，并引入过渡感知记忆来解决这些问题。
- 在RoboMemArena基准测试中，BATON的任务成功率提高了11.6%，累积成功率提高了14.9%，显示出显著的性能提升。
📝 摘要（中文）
长时间机器人操作将多个接触丰富的技能链成一个多阶段任务。尽管视觉-语言-动作（VLA）模型在单个技能上取得了进展，但在任务链中仍然存在错误累积的问题。BATON通过将子任务作为探索单元，采用短时间范围内的探索策略，并引入过渡感知记忆，显著提高了任务成功率。实验结果表明，BATON在长时间基准RoboMemArena上比现有最优方法提高了11.6%的任务成功率和14.9%的累积成功率。
🔬 方法详解
问题定义：论文要解决的问题是长时间机器人操作中任务链的错误累积和子任务之间的约束。现有方法在多阶段任务中，错误会在各个阶段之间传播，导致难以定位失败原因。
核心思路：BATON的核心思路是将每个子任务作为独立的探索单元，在短时间范围内进行探索并存储解决方案，从而降低探索成本并提高失败定位的准确性。
技术框架：BATON的整体架构包括子任务探索、过渡感知记忆和验证代理。每个子任务在短时间内被探索，解决方案被存储，长时间轨迹由这些解决方案组合而成。
关键创新：BATON的主要创新在于引入了过渡感知记忆，使得在子任务之间的过渡更加顺畅，确保了子任务的成功状态能够被后续任务有效利用。
关键设计：BATON在设计上不更新参数，而是通过验证代理控制VLA的调用时机，确保在场景准备好后才进行调用。同时，设计了交接过渡和前瞻过渡，以恢复被前一个子任务影响的状态。
🖼️ 关键图片
📊 实验亮点
BATON在长时间基准RoboMemArena上的实验结果显示，任务成功率提高了11.6%，累积成功率提高了14.9%。这些结果表明BATON在处理复杂多阶段任务时的有效性，超越了现有最优方法。
🎯 应用场景
该研究的潜在应用领域包括工业机器人、服务机器人和自动化生产线等。通过提高长时间操作的成功率，BATON能够显著提升机器人在复杂环境中的操作效率和可靠性，具有重要的实际价值和广泛的应用前景。
📄 摘要（原文）
Long-horizon robot manipulation chains many contact-rich skills into one multi-stage task. Vision-language-action (VLA) models increasingly master the individual skills, yet the chain still fails: errors compound beyond the policy's ability to correct, and one subtask silently constrains the next. A promising recipe freezes the VLA and puts an LLM agent in charge: it plans in language, moves in free space with analytic primitives, invokes the VLA only for contact-rich segments, and writes adaptation into language memory. Applied to long horizons, it breaks twice. (1) Competence comes from whole-task exploration at test time, whose cost is multiplicative in stages: if one stage needs T episodes, a K-stage task needs about T^K, and a failure does not reveal which stage caused it. (2) It has no representation of transitions: the VLA primitive carries an exit but no entry condition, so a subtask can succeed in a form its successor cannot use. We present BATON. Against (1), BATON makes the subtask the unit of exploration: each is explored in the cheap short-horizon regime and its solution stored in memory; a long-horizon trajectory is then composed from these solutions rather than discovered whole. Cost becomes additive (T*K) and every failure is attributed to a single stage. Against (2), BATON equips exploration with a transition-aware memory. Within a subtask, a verifier agent governs the invocation transition: the VLA is called only after the wrist view confirms the scene is ready. Across subtasks, a handoff transition restores an entry state disturbed by the predecessor's residue, and a lookahead transition selects the strategy whose outcome the successor can inherit. No parameters are updated. On the long-horizon benchmark RoboMemArena, BATON improves task success by 11.6% and cumulative success by 14.9% over the SoTA.