Conformal Reachability for Safe Control in Unknown Environments

Designing provably safe control is a core problem in trustworthy autonomy. However, most prior work in this regard assumes either that the system dynamics are known or deterministic, or that the state and action space are finite, significantly limiting application scope. We address this limitation by developing a probabilistic verification framework for unknown dynamical systems which combines conformal prediction with reachability analysis. In particular, we use conformal prediction to obtain valid uncertainty intervals for the unknown dynamics at each time step, with reachability then verifying whether safety is maintained within the conformal uncertainty bounds. Next, we develop an algorithmic approach for training control policies that optimize nominal reward while also maximizing the planning horizon with sound probabilistic safety guarantees. We evaluate the proposed approach in seven safe control settings spanning four domains – cartpole, lane following, drone control, and safe navigation – for both affine and nonlinear safety specifications. Our experiments show that the policies we learn achieve the strongest provable safety guarantees while still maintaining high average reward.

💡 Research Summary

The paper tackles the fundamental problem of guaranteeing safety for autonomous agents operating in environments where the system dynamics are unknown or stochastic. Traditional safe reinforcement‑learning approaches either assume known, deterministic dynamics or rely on finite state/action spaces, which severely limits their applicability. To overcome these constraints, the authors propose a probabilistic verification framework that fuses conformal prediction with finite‑horizon reachability analysis, termed Conformal Safety Analysis (CSA).

First, a neural‑network dynamics model (\hat f_\beta) is trained on collected trajectories. Using split conformal prediction, the method constructs valid prediction intervals (