Safety Filter Design for Neural Network Systems via Convex Optimization
This work addresses safety-critical control problems for neural network-based systems, which is incremental as it builds on existing verification and MPC methods.
The paper tackles the challenge of synthesizing provably safe controllers for neural network systems by proposing a safety filter based on convex optimization, which ensures robust constraint satisfaction under disturbances, and demonstrates efficacy on a nonlinear pendulum system with numerical results.
With the increase in data availability, it has been widely demonstrated that neural networks (NN) can capture complex system dynamics precisely in a data-driven manner. However, the architectural complexity and nonlinearity of the NNs make it challenging to synthesize a provably safe controller. In this work, we propose a novel safety filter that relies on convex optimization to ensure safety for a NN system, subject to additive disturbances that are capable of capturing modeling errors. Our approach leverages tools from NN verification to over-approximate NN dynamics with a set of linear bounds, followed by an application of robust linear MPC to search for controllers that can guarantee robust constraint satisfaction. We demonstrate the efficacy of the proposed framework numerically on a nonlinear pendulum system.