发表机构
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences; School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院自动化研究所多模态人工智能系统国家重点实验室; 中国科学院大学人工智能学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有神经形态触觉系统依赖主机预处理的问题，提出GelNeuro感算融合系统，搭配硬件感知权重钳位策略，实现低功耗高准确率的端侧纹理识别。
AI 中文摘要
神经形态视觉-触觉传感为低延迟、低功耗机器人感知提供了极具前景的范式。但现有系统在芯片推理前仍严重依赖主机完成事件读取、预处理或转发工作。本文提出GelNeuro，一款全集成感算融合视觉-触觉系统，将基于GelSight Mini的光学触觉前端直接与Speck2f神经形态片上系统（SoC）配对。接触引发的标记运动将被捕获为动态视觉传感器（DVS）事件，通过片上网络路由至脉冲卷积神经网络（SCNN）分类器。为缓解8位部署过程中的精度下降问题，本文提出了一种硬件感知权重钳位策略。在15类自然纹理识别任务上开展评估，实体芯片的硬件在环测试在80ms推理窗口内实现了96.3%的准确率。值得注意的是，该系统的板级有源功耗仅为19.6mW，比同基准下的传统CPU/GPU基线低三个数量级以上。GelNeuro在未见过的接触深度下也展现出鲁棒的泛化能力，证明了边缘神经形态硬件上直接传感器到芯片的触觉识别方案的可行性。
英文摘要
Neuromorphic visuo-tactile sensing offers a promising paradigm for low-latency and low-power robotic perception. However, existing systems still rely heavily on a host computer for event readout, preprocessing, or relaying prior to chip inference. This paper presents GelNeuro, a fully integrated sensing-computing visuo-tactile system that directly pairs a GelSight Mini-based optical tactile front end with the Speck2f neuromorphic system-on-chip (SoC). Contact-induced marker motions are captured as dynamic vision sensor (DVS) events and routed through the on-chip network to a spiking convolutional neural network (SCNN) classifier. To mitigate accuracy degradation during 8-bit deployment, a hardware-aware weight clamping strategy is introduced. Evaluated on a 15-class natural texture recognition task, hardware-in-the-loop testing on the physical chip achieves a 96.3% accuracy within an 80 ms inference window. Notably, the system consumes only 19.6 mW of board-level active power-over three orders of magnitude lower than conventional CPU/GPU baselines on the same benchmark. GelNeuro also exhibits robust generalization across unseen contact depths, demonstrating the viability of direct sensor-to-chip tactile recognition on edge neuromorphic hardware.
CommentsThe authors withdraw this preprint as the work requires substantial revision and additional validation. The current version is not suitable for public dissemination, and further development of the research is needed
