Stochastic model predictive control of nonlinear continuous-time systems using the unscented transformation

The paper presents a stochastic model predictive control (MPC) scheme for nonlinear continuous-time stochastic systems with chance constraints. The approach uses the unscented transformation to predict the mean and covariance of the nonlinear continuous-time dynamics. Special emphasis is put on integrating the resulting unscented MPC formulation within a real-time gradient algorithm by exploiting the structure of the optimality conditions. The effectiveness of the approach is demonstrated for an automotive emergency braking scenario with collision avoidance.