Learning Neural Control Barrier Functions from Expert Demonstrations using Inverse Constraint Learning

2025-10-24

AI Summary

This work addresses the safety control problem in autonomous systems where failure states cannot be explicitly defined. We propose an end-to-end framework integrating inverse constraint learning with neural control barrier functions (CBFs). Unlike conventional approaches, our method requires no manually annotated safe/unsafe labels; instead, it infers implicit safety constraints directly from expert demonstration trajectories, automatically generating state-wise safety labels and training a differentiable deep neural network that provably satisfies CBF conditions. Our key contributions are: (i) the first application of inverse constraint learning to *unsupervised* construction of neural CBFs, enabling automatic discovery of safety-invariant sets; and (ii) elimination of reliance on prior knowledge or explicit modeling of failure sets—traditionally required for CBF synthesis. Evaluated on four simulation benchmarks, our method achieves performance comparable to supervised baselines trained on ground-truth safety labels, and substantially outperforms existing unsupervised methods, demonstrating strong generalization and practical applicability.

Technology Category

• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Intelligent Robots: Behavior Learning & Control

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Safety is a fundamental requirement for autonomous systems operating in critical domains. Control barrier functions (CBFs) have been used to design safety filters that minimally alter nominal controls for such systems to maintain their safety. Learning neural CBFs has been proposed as a data-driven alternative for their computationally expensive optimization-based synthesis. However, it is often the case that the failure set of states that should be avoided is non-obvious or hard to specify formally, e.g., tailgating in autonomous driving, while a set of expert demonstrations that achieve the task and avoid the failure set is easier to generate. We use ICL to train a constraint function that classifies the states of the system under consideration to safe, i.e., belong to a controlled forward invariant set that is disjoint from the unspecified failure set, and unsafe ones, i.e., belong to the complement of that set. We then use that function to label a new set of simulated trajectories to train our neural CBF. We empirically evaluate our approach in four different environments, demonstrating that it outperforms existing baselines and achieves comparable performance to a neural CBF trained with the same data but annotated with ground-truth safety labels.

Problem

• Learning neural control barrier functions from expert demonstrations
• Handling unspecified failure states in autonomous systems
• Using inverse constraint learning for safety classification

Innovation

• Learning neural CBFs via inverse constraint learning
• Training constraint functions from expert demonstrations
• Generating safety labels from simulated trajectories