面向可信自主机器人：一种基于可解释人工智能的决策框架
Towards Trustworthy Autonomous Robots: An Explainable AI-Based Decision Framework
查看机构详情
- Hezarfen LLC（赫扎芬有限责任公司）
- Grand Canyon University（大峡谷大学）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
针对自主机器人决策不可审计问题，提出基于可解释AI的TRACE框架，经500次模拟实验，其在三项指标上表现优异，符合欧盟AI法案要求。
中文摘要 AI 辅助
由深度学习驱动的自主机器人面临一个根本性的可审计性挑战：当事故发生时，调查人员无法重现系统做出特定决策的原因。本文提出了TRACE（可信执行透明推理架构），这是一种决策框架，可确保每一项自主动作都能通过记录的因果链追溯到传感器证据。该框架将决策制定组织为四个可审计层：用于基于证据的实体识别的语义感知、用于带因果图的概率状态估计的信念推理、用于带反事实记录的约束感知规划的动作合成，以及用于合规性监控的执行验证。TRACE与模型无关，旨在集成基于学习的感知模块（CNN、Transformer），同时保留决策级可审计性。我们使用三个客观指标评估该框架：证据可追溯性（传感器到决策的链接）、决策可重现性（事后分析能力）和时间连续性（审计线索完整性）。对仓库机器人导航的实验评估表明，在500个模拟决策周期中，TRACE实现了98.6%的证据可追溯性、99.0%的时间连续性和98.1%的决策可重现性。事后方法如LIME提供特征归因，但缺乏决策级重现所需的人工结构。该框架满足欧盟AI法案对高风险系统透明度的要求，并为安全关键型自主系统的可解释人工智能做出贡献。
英文摘要
Autonomous robots powered by deep learning face a fundamental auditability challenge: when incidents occur, investigators cannot reconstruct why the system made specific decisions. This paper presents TRACE (Transparent Reasoning Architecture for Credible Execution), a decision framework that ensures every autonomous action can be traced back to sensor evidence through documented causal chains. The framework organizes decision-making into four auditable layers: Semantic Perception for evidence-grounded entity recognition, Belief Reasoning for probabilistic state estimation with causal graphs, Action Synthesis for constraint-aware planning with counterfactual documentation, and Execution Verification for compliance monitoring. TRACE is model-agnostic yet designed to integrate learning-based perception modules (CNNs, transformers) while preserving decision-level auditability. We evaluate the framework using three objective metrics: Evidence Traceability (sensor-to-decision linkage), Decision Reconstructability (post-hoc analysis capability), and Temporal Continuity (audit trail completeness). Experimental evaluation on warehouse robot navigation demonstrates that TRACE achieves 98.6% evidence traceability, 99.0% temporal continuity, and 98.1% decision reconstructability across 500 simulated decision cycles. Post-hoc methods like LIME provide feature attributions but lack the artifact structure needed for decision-level reconstruction. The framework addresses EU AI Act requirements for high-risk system transparency and contributes to Explainable AI for safety-critical autonomous systems.
