Self-Adaptive Anomaly Detection with Reinforcement Learning and Human Feedback in Connected Vehicles
arXiv:2607.08373v1 Announce Type: new Abstract: Connected vehicles are autonomous cyber-physical systems whose behavior must be continuously monitored during operation to detect deviations from normal operation before they propagate into failures. Such evaluation is challenging because the systems themselves evolve: over-the-air updates, configuration changes, and shifting workloads alter the definition of normal behavior, causing static diagnostic methods to degrade silently over time. Existing approaches typically address either automated model adaptation or operator integration in isolation
The increasing complexity and autonomy of cyber-physical systems like connected vehicles necessitate more robust, adaptive anomaly detection methods to ensure continuous reliability.
This research addresses a critical vulnerability in autonomous systems, where traditional diagnostic methods fail due to dynamic operating environments and frequent updates, leading to silent degradation and potential failures.
The integration of reinforcement learning and human feedback allows for self-adaptive anomaly detection in evolving systems, shifting from static monitoring to dynamic, learning-based diagnostics.
- · Automotive Manufacturers
- · AI/ML Software Providers
- · Connected Vehicle Operators
- · Cybersecurity Firms
- · Providers of Static Diagnostic Tools
- · Traditional Anomaly Detection Methods
More reliable and safer connected vehicle operations, reducing maintenance costs and increasing public trust.
Expansion of similar self-adaptive AI-driven monitoring systems into other critical infrastructure and cyber-physical domains.
Accelerated development of truly autonomous systems with higher levels of resilience and self-diagnosing capabilities, reducing human intervention requirements.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG