Skip to content

论文arXiv 2610.1904413 天前

RingNav:10.6k参数视觉航向估计器实现Pi Zero 27Hz实时推理 ​

本文由 AI 根据公开内容整理,可能有误,请以原文为准:RingNav: Task-Shaped Circular Memory for Visual Heading(arXiv 2610.19044)

核心内容 ​

针对微型机器人计算与存储资源极度受限的工程挑战,研究团队发布RingNav视觉航向估计系统。该模型仅含10.6k参数,摒弃传统冗余状态维护,将持久化记忆严格约束于圆形流形S^1。在TartanAir Visual SLAM Challenge基准测试中,相比同规模学习型记忆模型,其平均绝对转向误差降低27%,4秒累积航向误差减少15%,且具备零样本迁移至真实无人机平台的能力,在严重视觉退化环境下仍保持有效估计。

技术架构上,RingNav仅学习连续图像间的相对转向量,航向直接在圆环上合成,结构化更新机制保证了相同帧输入时转向为零及序列反转时的精确抵消。端侧部署性能优异,在Raspberry Pi Zero 2W上运行频率达27.3Hz,在超低功耗GAP8处理器上可达33.4Hz。这种极轻量级设计验证了任务导向型记忆结构在具身智能边缘计算中的可行性,为微型移动机器人提供了无需GPU的高频姿态感知方案。

关键看点 ​

  • 极致轻量化:模型仅10.6k参数,将记忆约束于S1流形,大幅降低微型机器人视觉导航算力门槛
  • 精度提升:TartanAir数据集平均绝对转向误差降27%,4秒累积航向误差减15%,优于同尺寸模型
  • 端侧高频推理:Raspberry Pi Zero 2W达27.3Hz,超低功耗GAP8处理器达33.4Hz,满足实时控制需求
  • 结构化鲁棒性:具备相同帧零转向与序列反转精确抵消特性,支持零样本迁移至真实无人机平台

工程与产业观察 ​

RingNav证明了在微瓦级算力平台上,通过拓扑约束替代暴力堆叠参数可实现高效感知。其S1流形记忆设计解决了小模型漂移问题,为昆虫级机器人、微型无人机及大规模集群提供了可落地的纯视觉航向估计范式,对边缘端具身智能算法压缩具有重要参考价值。

英文摘要 (Abstract) ​

Tiny robots have limited computation and memory, yet visual navigation systems often maintain more state than a task requires. We ask whether persistent memory can instead be shaped by the task. We present \textbf{RingNav}, a (10.6)k--parameter visual heading estimator that constrains its memory to the circular manifold (S^1). RingNav learns only the turn between consecutive images, while heading is composed directly on the circle. On the TartanAir Visual SLAM Challenge, RingNav reduces mean absolute turn error by (27%) and accumulated heading error over $4$ sec by (15%) relative to a similarly sized learned-memory model. The structured update also gives exactly zero turn for identical frames and exact cancellation under sequence reversal. Without fine-tuning, RingNav transfers to real-world drone experiments and continues estimating heading under severe visual degradation. RingNav runs at $27.3$Hz on Raspberry Pi Zero 2W and can be deployed on an ultra-low power GAP8 processor at $33.4$Hz.

作者:Naitri Rajyaguru, Aryan Tajne, Stephen Jardim, Cornelia Fermuller, Yiannis Aloimonos

来源

本页事实对应的公开页面,以原文为准。

  1. arXiv 2610.19044:RingNav: Task-Shaped Circular Memory for Visual Heading访问 2026-10-16