BarrierNet: A Safety-Guaranteed Layer for Neural Networks
This addresses safety constraints for neural controllers in dynamic environments, representing an incremental improvement by adapting existing CBF methods to be more flexible and trainable.
The paper tackled the problem of overly conservative control barrier functions (CBFs) in neural network-based controllers by introducing differentiable higher-order CBFs that are end-to-end trainable, softening definitions to reduce conservativeness while maintaining safety guarantees, and demonstrated effectiveness in control problems like traffic merging and robot navigation.
This paper introduces differentiable higher-order control barrier functions (CBF) that are end-to-end trainable together with learning systems. CBFs are usually overly conservative, while guaranteeing safety. Here, we address their conservativeness by softening their definitions using environmental dependencies without loosing safety guarantees, and embed them into differentiable quadratic programs. These novel safety layers, termed a BarrierNet, can be used in conjunction with any neural network-based controller, and can be trained by gradient descent. BarrierNet allows the safety constraints of a neural controller be adaptable to changing environments. We evaluate them on a series of control problems such as traffic merging and robot navigations in 2D and 3D space, and demonstrate their effectiveness compared to state-of-the-art approaches.