Risk-Aware Maximum Hands-Off Control Using Worst-Case Conditional Value-At-Risk

With the view of risks, this paper deals with problems maximum hands-off control which aims at minimizing length nonzero input. More specifically, we consider stochastic systems and seek sparse inputs that bring system state to a ball centered origin, such expected value states are further than given threshold from origin is small, thus risk outside ball. To deal problem, employ worst-case Conditional Value-at-Risk (CVaR) under assumption first two moments disturbance distribution known. In particular, kinds risk-aware problems: one enhances sparsity within threshold, other minimizes subject constraint. We also derive risk-constrained model predictive provide numerical example shows effectiveness proposed approach in networked systems.