统一感知验证与评估，实现动态环境下的可靠安全保证
Towards Unified Probabilistic Verification and Validation of Vision-Based Autonomy
- 用区间MDP融合感知验证与离线评估，适应分布外测试条件
- 在合成感知马尔可夫链和山地小车基准上实现严格的安全边界
- 适合关注自动驾驶感知可靠性与形式化验证的研究者
精准全面的情境感知是现代自主系统的关键能力。基于深度神经网络的视觉感知虽已广泛应用，但其黑箱特性及对环境不确定性与分布偏移的敏感性，使得形式化验证困难。现有的基于抽象的视觉自主验证方法依赖于严格的假设，如误差有界或分布已知，这些限制性假设削弱了安全保证的有效性，尤其在多样且不确定的运行环境中。本文提出一种统一感知验证与离线评估的方法，利用区间马尔可夫决策过程（interval MDPs），提供可直接响应分布外测试条件的灵活端到端安全保证。我们在一个具有明确定义状态估计分布的合成感知马尔可夫链和山地小车基准上进行了评估，结果表明，该方法能够对系统整体安全性给出紧致且严谨的边界约束。
原文摘要 · Abstract (English)
Precise and comprehensive situational awareness is a critical capability of modern autonomous systems. Deep neural networks that perceive task-critical details from rich sensory signals have become ubiquitous; however, their black-box behavior and sensitivity to environmental uncertainty and distribution shifts make them challenging to verify formally. Abstraction-based verification techniques for vision-based autonomy produce safety guarantees contingent on rigid assumptions, such as bounded errors or known unique distributions. Such overly restrictive and inflexible assumptions limit the validity of the guarantees, especially in diverse and uncertain test-time environments. We propose a methodology that unifies the verification models of perception with their offline validation. Our methodology leverages interval MDPs and provides a flexible end-to-end guarantee that adapts directly to the out-of-distribution test-time conditions. We evaluate our methodology on a synthetic perception Markov chain with well-defined state estimation distributions and a mountain car benchmark. Our findings reveal that we can guarantee tight yet rigorous bounds on overall system safety.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。