面向资源受限空间机器人的硬件加速实例分割方法及临界性分析

中文摘要 AI 辅助

自主月球任务需要在三个耦合约束下实现实时感知：极端弱光条件、有限星载计算资源，以及可能悄悄破坏推理过程的辐射诱发硬件故障。本文提出一种面向部署的、用于资源受限月球机器人的实例分割框架，在严格计算约束下同时解决量化校准和系统级故障暴露问题。首先，我们引入激活方差信息采样（Activation Variance Informative Sampling, AVIS），这是一种无标签校准策略，基于激活方差统计确定性选择校准样本。其次，我们将基于YOLO的分割模型部署在深度学习处理器单元（Deep Learning Processor Unit, DPU）上，通过架构修改减少CPU fallback路径，实现弱光条件下具有受限延迟的静态编译执行。我们进一步引入软件级临界性分析，以估计故障暴露并指导辐射受限操作下的缓解措施。在月球微型漫游车平台上，带偏差校正的AVIS恢复了69.8%的量化诱导精度损失，同时实现309毫秒推理延迟和5.7瓦功耗；针对性缓解措施将全局临界性降低了31.7%。结果证明了一种集成方法，为空间部署约束下的可靠、安全AI感知框架提供了蓝图。

英文摘要

Autonomous lunar missions require real-time per- ception under three coupled constraints: extreme low-light conditions, limited onboard compute, and radiation-induced hardware faults that can silently corrupt inference. We present a deployment-oriented instance segmentation framework for resource-constrained lunar robotics that jointly addresses quan- tization calibration and system-level fault exposure under strict compute constraints. First, we introduce Activation Variance Informative Sampling (AVIS), a label-free calibration strategy that deterministically selects calibration samples based on activation variance statistics. Second, we deploy a YOLO-based segmentation model on a Deep Learning Processor Unit (DPU) with architectural modifications that reduce CPU fallback paths and enable statically compiled execution with bounded latency in low-lighting conditions. We further introduce a software-level criticality analysis to estimate fault exposure and guide mitigation under radiation-constrained operation. On a lunar micro-rover platform, AVIS with bias correction recovers 69.8% of quantization-induced accuracy loss while achieving 309 ms inference latency and 5.7 W power consumption. Targeted mitigation reduces global criticality by 31.7%. The results demonstrate an integrated approach and a blueprint for a reliable and safe AI perception framework under space deployment constraints.

发表机构

• German Research Center for Artificial Intelligence GmbH (DFKI)（德国人工智能研究中心）
• University of Bremen（不来梅大学）

机构由 AI 辅助整理，请以论文原文为准。