How to Train Your Neural Control Barrier Function: Learning Safety Filters for Complex Input-Constrained Systems

2023/10/24 by Oswin So, So, Oswin, Zachary Serlin +11 · 12 citations
Computer Science · Engineering · #Adversarial Robustness in Machine Learning #FOS: Computer and information sciences #FOS: Mathematics #Fault Detection and Control Systems #Optimization and Control (math.OC) #Robotics (cs.RO) #Smart Grid Security and Resilience

paper · pdf · doi:10.48550/arxiv.2310.15478

openalex publication_date 2023/10/24 · openalex created_date 2023/10/26 · openalex updated_date 2026/07/28

Abstract

Control barrier functions (CBF) have become popular as a safety filter to guarantee the safety of nonlinear dynamical systems for arbitrary inputs. However, it is difficult to construct functions that satisfy the CBF constraints for high relative degree systems with input constraints. To address these challenges, recent work has explored learning CBFs using neural networks via neural CBF (NCBF). However, such methods face difficulties when scaling to higher dimensional systems under input constraints. In this work, we first identify challenges that NCBFs face during training. Next, to address these challenges, we propose policy neural CBF (PNCBF), a method of constructing CBFs by learning the value function of a nominal policy, and show that the value function of the maximum-over-time cost is a CBF. We demonstrate the effectiveness of our method in simulation on a variety of systems ranging from toy linear systems to an F-16 jet with a 16-dimensional state space. Finally, we validate our approach on a two-agent quadcopter system on hardware under tight input constraints.

Cited by

• Extracting Exact Lie Derivatives Without Backpropagation: A Dual Compiler for Neural Control Barrier Functions
• V-OCBF: Learning Safety Filters from Offline Data via Value-Guided Offline Control Barrier Functions
• Learning Control Barrier Functions with Deterministic Safety Guarantees
• How to Train Your Latent Control Barrier Function: Smooth Safety Filtering Under Hard-to-Model Constraints
• LEARN: Learning End-to-End Aerial Resource-Constrained Multi-Robot Navigation
• Learning Neural Control Barrier Functions from Expert Demonstrations using Inverse Constraint Learning
• Recurrent Control Barrier Functions: A Path Towards Nonparametric Safety Verification
• Safe and Stable Control via Lyapunov-Guided Diffusion Models
• From Space to Time: Enabling Adaptive Safety with Learned Value Functions via Disturbance Recasting
• EigenSafe: A Spectral Framework for Learning-Based Probabilistic Safety Assessment
• ORN-CBF: Learning Observation-conditioned Residual Neural Control Barrier Functions via Hypernetworks
• Safe Gap-based Planning in Dynamic Settings

Related

• Safe Control Under Input Limits with Neural Control Barrier Functions
• Learning Performance-Oriented Control Barrier Functions Under Complex Safety Constraints and Limited Actuation
• Learning a Better Control Barrier Function
• Learning a Formally Verified Control Barrier Function in Stochastic Environment
• Learning Neural Control Barrier Functions from Expert Demonstrations using Inverse Constraint Learning