Certifiable Neural Control for Safe Autonomy and Robotics

Safety is central to autonomous systems and robots since a single failure could lead to catastrophic results. In unstructured complex environments where system states and environment information are not available, the safety-critical control problem is much more challenging. In this talk, I will first discuss safety from a control theoretic perspective with Control Barrier Functions (CBFs). CBFs capture the evolution of the safety requirements during the execution of a control system and can be used to guarantee safety for all times due to their forward invariance. Next, this talk will introduce an approach for extending the use of CBFs to machine learning-based control, using differentiable CBFs that are end-to-end trainable and adaptively guarantee safety using environmental dependencies. These novel safety layers give rise to new neural network (NN) architectures such as what we have termed BarrierNet.  In machine learning and robot learning, the interpretability of a NN is crucial. The talk will further introduce a novel method called invariance propagation through the NN. This approach enables causal reasoning of the NN's parameters or inputs with respect to robot behaviors, as well as introducing guarantees. Finally, I will show how we can certify more powerful generative AI, such as diffusion models, for generalizable and safe autonomy and robotics. These techniques have been successfully applied to various robotic systems, such as autonomous ground vehicles, surface vessels, and flight vehicles, legged robots, robot swarms, soft robots, and manipulators.