Joint Learning of Feasibility-Aware Signal Temporal Logic and BarrierNet for Robust and Correct Control

2025-12-07

AI Summary

Existing CBF-STL integration methods suffer from conservative behavior, infeasibility under tight input constraints, and difficulty in ensuring long-horizon STL task satisfaction due to fixed hyperparameters and myopic optimization. This paper proposes an end-to-end differentiable framework unifying learnable high-order control barrier functions (HOCBFs) with signal temporal logic (STL). Its core contributions are: (1) a feasibility-aware time-varying HOCBF mechanism incorporating a unified robustness metric; (2) a tripartite neural architecture—InitNet, RefNet, and BarrierNet—that jointly learns and dynamically adapts constraint parameters, eliminating manual tuning and resolving optimization infeasibility; and (3) real-time adaptive control via differentiable quadratic programming. Experiments demonstrate substantial improvements in STL robustness and long-horizon task completion rates under stringent input constraints, while rigorously guaranteeing controller feasibility and STL task consistency.

Technology Category

• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Search and Optimization: Learning to Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Control Barrier Functions (CBFs) have emerged as a powerful tool for enforcing safety in optimization-based controllers, and their integration with Signal Temporal Logic (STL) has enabled the specification-driven synthesis of complex robotic behaviors. However, existing CBF-STL approaches typically rely on fixed hyperparameters and myopic, per-time step optimization, which can lead to overly conservative behavior, infeasibility near tight input limits, and difficulty satisfying long-horizon STL tasks. To address these limitations, we propose a feasibility-aware learning framework that embeds trainable, time-varying High Order Control Barrier Functions (HOCBFs) into a differentiable Quadratic Program (dQP). Our approach provides a systematic procedure for constructing time-varying HOCBF constraints for a broad fragment of STL and introduces a unified robustness measure that jointly captures STL satisfaction, QP feasibility, and control-bound compliance. Three neural networks-InitNet, RefNet, and an extended BarrierNet-collaborate to generate reference inputs and adapt constraint-related hyperparameters automatically over time and across initial conditions, reducing conservativeness while maximizing robustness. The resulting controller achieves STL satisfaction with strictly feasible dQPs and requires no manual tuning. Simulation results demonstrate that the proposed framework maintains high STL robustness under tight input bounds and significantly outperforms fixed-parameter and non-adaptive baselines in complex environments.

Problem

• Addresses overly conservative behavior in CBF-STL control systems.
• Solves infeasibility issues near tight input limits.
• Enhances satisfaction of long-horizon STL tasks.

Innovation

• Feasibility-aware learning with trainable time-varying HOCBFs
• Differentiable QP embedding for systematic STL constraint construction
• Neural networks adapt hyperparameters to reduce conservativeness