Gated Uncertainty-Aware Runtime Dual Invariants for Neural Signal-Controlled Robotics

2025-11-25

AI Summary

To address the insufficient reliability and trustworthiness of neural signal–controlled assistive systems in safety-critical applications, this paper proposes a real-time neurosymbolic verification framework. The framework integrates uncertainty-aware decoding, symbolic goal grounding, gated runtime verification, and lightweight neural networks to establish a two-tier runtime monitoring and confidence calibration mechanism: an upper tier employs dual-invariant monitoring for progressive response to signal degradation, while a lower tier enhances decision interpretability and safety via auditable intent–plan–action tracing. The system achieves sub-millisecond decision latency at 100 Hz sampling. Evaluated on the BNCI2014 dataset, it attains safety rates of 94–97% despite low decoding accuracies (27–46%), and improves intervention correctness by 1.7× under noise—demonstrating high robustness and practical feasibility.

Technology Category

• Machine Learning: Neuro-Symbolic Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Natural Language Processing: Safety and Robustness

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Safety-critical assistive systems that directly decode user intent from neural signals require rigorous guarantees of reliability and trust. We present GUARDIAN (Gated Uncertainty-Aware Runtime Dual Invariants), a framework for real-time neuro-symbolic verification for neural signal-controlled robotics. GUARDIAN enforces both logical safety and physiological trust by coupling confidence-calibrated brain signal decoding with symbolic goal grounding and dual-layer runtime monitoring. On the BNCI2014 motor imagery electroencephalogram (EEG) dataset with 9 subjects and 5,184 trials, the system performs at a high safety rate of 94-97% even with lightweight decoder architectures with low test accuracies (27-46%) and high ECE confidence miscalibration (0.22-0.41). We demonstrate 1.7x correct interventions in simulated noise testing versus at baseline. The monitor operates at 100Hz and sub-millisecond decision latency, making it practically viable for closed-loop neural signal-based systems. Across 21 ablation results, GUARDIAN exhibits a graduated response to signal degradation, and produces auditable traces from intent, plan to action, helping to link neural evidence to verifiable robot action.

Problem

• Ensuring reliability in neural signal-controlled robotics systems
• Providing real-time safety verification for brain-controlled assistive devices
• Maintaining system trustworthiness under low decoder accuracy conditions

Innovation

• Gated uncertainty-aware runtime dual invariants framework
• Couples confidence-calibrated decoding with symbolic grounding
• Provides real-time neuro-symbolic verification for neural robotics