Safety Certification in the Latent space using Control Barrier Functions and World Models

2025/07/18 by Mehul Anand, Anand, Mehul, Shishir Kolathaya +1 · 1 citation
Computer Science · Engineering · #Adversarial Robustness in Machine Learning #Autonomous Vehicle Technology and Safety #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG) #Robotics (cs.RO) #Systems and Control (eess.SY) #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2507.13871

openalex publication_date 2025/07/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Synthesising safe controllers from visual data typically requires extensive supervised labelling of safety-critical data, which is often impractical in real-world settings. Recent advances in world models enable reliable prediction in latent spaces, opening new avenues for scalable and data-efficient safe control. In this work, we introduce a semi-supervised framework that leverages control barrier certificates (CBCs) learned in the latent space of a world model to synthesise safe visuomotor policies. Our approach jointly learns a neural barrier function and a safe controller using limited labelled data, while exploiting the predictive power of modern vision transformers for latent dynamics modelling.

Citations

• CP-NCBF: A Conformal Prediction-based Approach to Synthesize Verified Neural Control Barrier Functions
• A Physics-Informed Machine Learning Framework for Safe and Optimal Control of Autonomous Systems
• Generalizing Safety Beyond Collision-Avoidance via Latent-Space Reachability Analysis
• DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning
• Semi-Supervised Safe Visuomotor Policy Synthesis using Barrier Certificates
• Neural Control Barrier Functions for Safe Navigation
• Exact Imposition of Safety Boundary Conditions in Neural Reachable Tubes
• Learning a Formally Verified Control Barrier Function in Stochastic Environment
• DINOv2: Learning Robust Visual Features without Supervision
• Control Barrier Functions in Dynamic UAVs for Kinematic Obstacle Avoidance: A Collision Cone Approach
• Mastering Diverse Domains through World Models
• Model-based Safe Deep Reinforcement Learning via a Constrained Proximal Policy Optimization Algorithm
• Enforcing safety for vision-based controllers via Control Barrier Functions and Neural Radiance Fields
• SABLAS: Learning Safe Control for Black-box Dynamical Systems
• Control Barrier Functions for Unknown Nonlinear Systems using Gaussian\n Processes
• Learning for Safety-Critical Control with Control Barrier Functions
• Control Barrier Functions: Theory and Applications
• Hamilton-Jacobi Reachability: A Brief Overview and Recent Advances
• Constrained Policy Optimization
• Safe Model-based Reinforcement Learning with Stability Guarantees

Cited by

• How to Train Your Latent Control Barrier Function: Smooth Safety Filtering Under Hard-to-Model Constraints

Related

• Semi-Supervised Safe Visuomotor Policy Synthesis using Barrier Certificates
• Safe Physics-Informed Machine Learning for Dynamics and Control
• Safety Considerations in Deep Control Policies with Safety Barrier Certificates Under Uncertainty
• What You Don't Know Can Hurt You: How Well do Latent Safety Filters Understand Partially Observable Safety Constraints?
• How Safe Am I Given What I See? Calibrated Prediction of Safety Chances for Image-Controlled Autonomy