Scalable Verification of Neural Control Barrier Functions Using Linear Bound Propagation
This addresses a key limitation in safety-critical control systems by enabling more efficient verification of neural CBFs, though it is an incremental improvement over existing verification techniques.
The paper tackles the computational bottleneck in verifying neural control barrier functions (CBFs) for safety certification by introducing a framework based on linear bound propagation and McCormick relaxation, achieving scalability to larger networks than state-of-the-art methods as shown in numerical experiments.
Control barrier functions (CBFs) are a popular tool for safety certification of nonlinear dynamical control systems. Recently, CBFs represented as neural networks have shown great promise due to their expressiveness and applicability to a broad class of dynamics and safety constraints. However, verifying that a trained neural network is indeed a valid CBF is a computational bottleneck that limits the size of the networks that can be used. To overcome this limitation, we present a novel framework for verifying neural CBFs based on piecewise linear upper and lower bounds on the conditions required for a neural network to be a CBF. Our approach is rooted in linear bound propagation (LBP) for neural networks, which we extend to compute bounds on the gradients of the network. Combined with McCormick relaxation, we derive linear upper and lower bounds on the CBF conditions, thereby eliminating the need for computationally expensive verification procedures. Our approach applies to arbitrary control-affine systems and a broad range of nonlinear activation functions. To reduce conservatism, we develop a parallelizable refinement strategy that adaptively refines the regions over which these bounds are computed. Our approach scales to larger neural networks than state-of-the-art verification procedures for CBFs, as demonstrated by our numerical experiments.