Safety Certification in the Latent space using Control Barrier Functions and World Models
This addresses the challenge of safety certification in real-world robotic or autonomous systems where labeling safety-critical data is impractical, though it appears incremental by combining existing techniques like control barrier functions and world models.
The paper tackles the problem of synthesizing safe controllers from visual data without extensive supervised labeling by introducing a semi-supervised framework that uses control barrier certificates in the latent space of a world model, resulting in scalable and data-efficient safe visuomotor policies.
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.