CoIn-SafeLink: Safety-critical Control With Cost-sensitive Incremental Random Vector Functional Link Network

2025-03-19

AI Summary

In safety-critical control under dynamically evolving, irregular hazard regions, analytically modeling the unsafe set remains challenging. Method: This paper proposes a Control Barrier Function (CBF) synthesis method based on a cost-sensitive incremental Random Vector Functional Link (RVFL) network. It enforces zero false negatives via a cost-sensitive learning mechanism and derives an incremental update theorem with closed-form gradient expressions to enable real-time, precise CBF adaptation as hazard boundaries evolve. Contribution/Results: Evaluated on a 3-DOF UAV attitude control task, the method accurately learns dynamic hazard boundaries while achieving approximately fivefold faster update speed than baseline approaches. It guarantees theoretical safety—via strict satisfaction of the CBF condition—and meets real-time implementation requirements.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Reasoning under Uncertainty: Sequential Decision Making
• Machine Learning: Causal Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Control barrier functions (CBFs) play a crucial role in achieving the safety-critical control of robotic systems theoretically. However, most existing methods rely on the analytical expressions of unsafe state regions, which is often impractical for irregular and dynamic unsafe regions. In this paper, a novel CBF construction approach, called CoIn-SafeLink, is proposed based on cost-sensitive incremental random vector functional-link (RVFL) neural networks. By designing an appropriate cost function, CoIn-SafeLink achieves differentiated sensitivities to safe and unsafe samples, effectively achieving zero false-negative risk in unsafe sample classification. Additionally, an incremental update theorem for CoIn-SafeLink is proposed, enabling precise adjustments in response to changes in the unsafe region. Finally, the gradient analytical expression of the CoIn-SafeLink is provided to calculate the control input. The proposed method is validated on a 3-degree-of-freedom drone attitude control system. Experimental results demonstrate that the method can effectively learn the unsafe region boundaries and rapidly adapt as these regions evolve, with an update speed approximately five times faster than comparison methods. The source code is available at https://github.com/songqiaohu/CoIn-SafeLink.

Problem

• Constructs CBFs for irregular dynamic unsafe regions
• Ensures zero false-negative risk in unsafe classification
• Enables rapid adaptation to evolving unsafe regions

Innovation

• Cost-sensitive RVFL neural networks for CBF construction
• Incremental update theorem for dynamic unsafe regions
• Gradient analytical expression for control input calculation