EvoEye: Self-Evolving Runtime Monitoring for Autonomous Driving Systems

• URL: http://arxiv.org/abs/2607.03755v2
• Date: Tue, 07 Jul 2026 09:44:26 GMT
• Title: EvoEye: Self-Evolving Runtime Monitoring for Autonomous Driving Systems
• Abstract summary: We propose EvoEye, which identifies the current monitor's errors, generates informative executions accordingly, and updates the monitor through self-evolution.<n>To enable effective self-evolution, EvoEye combines a capable runtime monitor with targeted scenario acquisition.<n> FusionMonitor improves frame-level Recall by up to 37.8 percentage points at a false positive rate of 0.05, with 2.49 ms latency and 2.8-4.2 seconds of median warning time.<n>BlindSpotEvolver outperforms uniform and violation-oriented sampling by up to 13.2 F1 points on previously missed unsafe contexts.
• Score: 15.753537830005596
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Runtime monitoring is essential for detecting impending hazards in autonomous driving systems (ADSs). However, existing ADS runtime monitors have fixed detection capabilities: rule-based monitors cover only manually specified hazards, while learning-based monitors depend heavily on their initial training data and may retain substantial prediction errors. We therefore propose EvoEye, which identifies the current monitor's errors, generates informative executions accordingly, and updates the monitor through self-evolution. To enable effective self-evolution, EvoEye combines a capable runtime monitor with targeted scenario acquisition. FusionMonitor learns cross-module temporal interactions for collision prediction, while BlindSpotEvolver converts current prediction errors into search guidance and uses density-aware mutation to acquire informative executions for subsequent monitor updates. We evaluate EvoEye on Baidu Apollo with CARLA in representative highway and urban scenarios. FusionMonitor improves frame-level Recall by up to 37.8 percentage points at a false positive rate of 0.05, with 2.49 ms latency and 2.8-4.2 seconds of median warning time. Under the same budget, BlindSpotEvolver outperforms uniform and violation-oriented sampling by up to 13.2 F1 points on previously missed unsafe contexts.

Related papers

• When the Martingale Never Stops Firing: Anytime-Valid Gating on Real Forecast Streams [0.614481021961242]
  Anytime-valid inference promises evidence that can be acted on at any moment.<n>A deployment inherits it only if the stream it monitors behaves exchangeably.
  arXiv  Detail & Related papers  (2026-08-31T09:32:11Z)
• Self-Adaptive Anomaly Detection with Reinforcement Learning and Human Feedback in Connected Vehicles [0.3324986723090369]
  This paper presents an online anomaly detection framework for autonomous CPS.<n>A factorized deep Q-network with self-attention selects the most suitable detector from a candidate pool for each monitored service.<n>A human-in-the-loop retraining mechanism, built around a pending transition buffer and a 60/40 prioritized replay strategy, allows the operator to incorporate expert knowledge.
  arXiv  Detail & Related papers  (2026-07-09T11:48:52Z)
• Risk-Aware Selective Multimodal Driver Monitoring with Driver-State World Modeling [13.540675152706589]
  Continuous driver monitoring in automated vehicles requires low-latency inference.<n>We propose a cost-aware selective inference framework for deployable multimodal driver monitoring.
  arXiv  Detail & Related papers  (2026-06-25T11:59:21Z)
• Do Activation Monitors Survive Model Updates? Benchmarking, Predicting, and Repairing Activation-Monitor Staleness [0.16921396880325779]
  We test whether activation monitors trained on a base model remain reliable after routine model updates.<n>We find a sharp split: quantization-style updates largely preserve frozen probe performance, while fine-tuning-style updates frequently make probes stale.<n>We also show that degradation is predictable from pre-deployment features, enabling revalidation budgets to be triaged toward the monitors most likely to fail.
  arXiv  Detail & Related papers  (2026-06-14T19:07:22Z)
• $D^2$-Monitor: Dynamic Safety Monitoring for Diffusion LLMs via Hesitation-Aware Routing [63.49501120848927]
  Safety monitoring for large language models (D-LLMs) remains largely unexplored.<n>We propose $D2$-Monitor, a bi-level safety monitor for D-LLMs.<n>$D2$-Monitor adopts a lightweight probe as an always-on monitor to jointly estimate hesitation and perform base classification.
  arXiv  Detail & Related papers  (2026-05-25T14:22:21Z)
• Improving Driver Drowsiness Detection via Personalized EAR/MAR Thresholds and CNN-Based Classification [0.0]
  Vision-based driver monitoring systems often rely on fixed Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) thresholds.<n>This paper proposes a personalized driver drowsiness detection system that monitors eyelid movements, head position, and yawning behavior in real time.<n>Deep learning-based Convolutional Neural Network (CNN) models are integrated to enhance accuracy in challenging scenarios.
  arXiv  Detail & Related papers  (2026-04-24T12:00:08Z)
• Detecting Object Tracking Failure via Sequential Hypothesis Testing [80.7891291021747]
  Real-time online object tracking in videos constitutes a core task in computer vision.<n>We propose interpreting object tracking as a sequential hypothesis test, wherein evidence for or against tracking failures is gradually accumulated over time.<n>We propose both supervised and unsupervised variants by leveraging either ground-truth or solely internal tracking information.
  arXiv  Detail & Related papers  (2026-02-13T14:57:15Z)
• Adaptive Attacks on Trusted Monitors Subvert AI Control Protocols [80.68060125494645]
  We study adaptive attacks by an untrusted model that knows the protocol and the monitor model.<n>We instantiate a simple adaptive attack vector by which the attacker embeds publicly known or zero-shot prompt injections in the model outputs.
  arXiv  Detail & Related papers  (2025-10-10T15:12:44Z)
• Temporal Misalignment Attacks against Multimodal Perception in Autonomous Driving [23.83773255227831]
  We introduce DejaVu, an attack that exploits the in-vehicular network and induces delays across sensor streams to create subtle temporal misalignments.<n>Our comprehensive attack analysis reveals the sensors' task-specific imbalanced sensitivities.<n>With a single-frame LiDAR delay, an attacker can reduce the car detection mAP by up to 88.5%, while with a three-frame camera delay, multiple object tracking accuracy (MOTA) for car drops by 73%.
  arXiv  Detail & Related papers  (2025-07-12T00:44:26Z)
• Natural Reflection Backdoor Attack on Vision Language Model for Autonomous Driving [55.96227460521096]
  Vision-Language Models (VLMs) have been integrated into autonomous driving systems to enhance reasoning capabilities.<n>We propose a natural reflection-based backdoor attack targeting VLM systems in autonomous driving scenarios.<n>Our findings uncover a new class of attacks that exploit the stringent real-time requirements of autonomous driving.
  arXiv  Detail & Related papers  (2025-05-09T20:28:17Z)
• Can we Defend Against the Unknown? An Empirical Study About Threshold Selection for Neural Network Monitoring [6.8734954619801885]
  runtime monitoring becomes essential to reject unsafe predictions during inference. Various techniques have emerged to establish rejection scores that maximize the separability between the distributions of safe and unsafe predictions. In real-world applications, an effective monitor also requires identifying a good threshold to transform these scores into meaningful binary decisions.
  arXiv  Detail & Related papers  (2024-05-14T14:32:58Z)
• Unsupervised Domain Adaptation for Self-Driving from Past Traversal Features [69.47588461101925]
  We propose a method to adapt 3D object detectors to new driving environments. Our approach enhances LiDAR-based detection models using spatial quantized historical features. Experiments on real-world datasets demonstrate significant improvements.
  arXiv  Detail & Related papers  (2023-09-21T15:00:31Z)
• Training-free Monocular 3D Event Detection System for Traffic Surveillance [93.65240041833319]
  Existing event detection systems are mostly learning-based and have achieved convincing performance when a large amount of training data is available. In real-world scenarios, collecting sufficient labeled training data is expensive and sometimes impossible. We propose a training-free monocular 3D event detection system for traffic surveillance.
  arXiv  Detail & Related papers  (2020-02-01T04:42:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.