TacMamba: 一种连接快速反射与慢速VLA推理的触觉历史压缩适配器
TacMamba: A Tactile History Compression Adapter Bridging Fast Reflexes and Slow VLA Reasoning
AI总结：
TacMamba通过触觉历史压缩器和双阶段训练策略，实现高频率触觉数据与低频视觉推理的高效融合，提升实时任务性能。
摘要：
在视觉模糊的操纵任务中，如检测按钮点击的触觉反馈通常是唯一的真实数据来源。然而，融合触觉数据面临显著挑战，因为存在时空不匹配：触觉感知需要高频率处理和长周期记忆（系统1），而视觉策略则以低控制频率运行（系统2）。现有架构难以弥合这一差距：Transformer对高频循环（>100Hz）计算上不可行，而LSTM在长交互历史中易遗忘。本文提出TacMamba，一种分层架构，将高带宽触觉反射与低频视觉规划对齐。我们的方法包含三个核心贡献：（1）一个定制的高频触觉接口，用于灵活集成；（2）基于Mamba的触觉历史压缩器，将连续力历史编码为紧凑状态，具有O(1)推理延迟（0.45 ms），使能够与VLA模型无缝融合，无需联合预训练；（3）一种触觉引导的双阶段训练策略，利用时间辨别进行自监督表示学习，并利用相等采样来缓解数据稀疏性。在离散计数和隐式状态切换的实验中，TacMamba实现了100%的成功率，显著优于视觉-only pi_0.5基线，同时严格满足硬实时约束。
英文摘要：
In visually ambiguous manipulation such as detecting button click tactile feedback is often the sole source of ground truth. However, fusing tactile data poses a significant challenge due to a spatiotemporal mismatch: tactile perception requires high-frequency processing with long-horizon memory (System 1), whereas visual policies operate at low control frequencies (System 2). Existing architectures struggle to bridge this gap: Transformers are computationally prohibitive for high-frequency loops (>100Hz), while LSTMs suffer from forgetting over extended interaction histories. In this paper, we introduce TacMamba, a hierarchical architecture that aligns high-bandwidth tactile reflexes with low-frequency visual planning. Our approach comprises three core contributions: (1) a custom high-frequency tactile interface designed for flexible integration; (2) a Mamba-based Tactile History Compressor that encodes continuous force history into a compact state with O(1) inference latency (0.45 ms), enabling plug-and-play fusion with VLA models without joint pre-training and (3) a Tactile-Guided Dual-Stage Training strategy that leverages temporal discrimination for self-supervised representation learning and phase-uniform sampling to mitigate data sparsity. Experiments on discrete counting and implicit state switching demonstrate that TacMamba achieves 100% success rates, significantly outperforming the visual-only pi_0.5 baseline, while strictly satisfying hard real-time constraints.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
