This paper addresses learning safe control laws from expert demonstrations. We assume that appropriate models of the system dynamics and the output measurement map are available, along with corresponding error bounds. We first propose robust output control barrier functions (ROCBFs) as a means to guarantee safety, as defined through controlled forward invariance of a safe set. We then present an optimization problem to learn ROCBFs from expert demonstrations that exhibit safe system behavior, e.g., data collected from a human operator. Along with the optimization problem, we provide verifiable conditions that guarantee validity of the obtained ROCBF. These conditions are stated in terms of the density of the data and on Lipschitz and boundedness constants of the learned function and the models of the system dynamics and the output measurement map. When the parametrization of the ROCBF is linear, then, under mild assumptions, the optimization problem is convex. We validate our findings in the autonomous driving simulator CARLA and show how to learn safe control laws from RGB camera images.

No abstract

The stress corrosion cracking (SCC) behavior of oxidedispersion-strengthened (ODS) 304 austenitic steels has been investigated in a chloride-rich aqueous environment at 143°C. ODS 304 alloys are found to be more resistant to SCC than the commercial AISI 304 steels. Under a constant tensile load of 177 MPa, the crack growth rate in ODS 304 steels is about one fourth of AISI 304 steels, and the time-to-failure of ODS 304 steels is 7.5 times of AISI 304 steels. Intergranular SCC dominates the fracture surface of AISI 304 steel, while in ODS 304 steel both intergranular and transgranular SCC occur. Electrochemical reactivity tests show ODS 304 steel is less sensitized than AISI 304, likely a result of a low carbon concentration and small grain size.

We take the first step in using vehicle-to-vehicle (V2V) communication to provide real-time onboard traffic predictions. In order to best utilize real-world V2V communication data, we integrate first principle models with deep learning. Specifically, we train recurrent neural networks to improve the predictions given by first principle models. Our approach is able to predict the velocity of individual vehicles up to a minute into the future with improved accuracy over first principlebased baselines. We conduct a comprehensive study to evaluate different methods of integrating first principle models with deep learning techniques. The source code for our models is available at https://github.com/Rose-STL-Lab/V2V-traffic-forecast.