A predictive guidance law with uncertain information about the target state

The paper presents a new terminal guidance law for pursuit-evasion problems with uncertain information about the target state. The law adopts a novel predictive guidance approach that explicitly employs the probability density of the predicted target position in the terminal constraint for the predictive control. The central idea in the formulation of this terminal constraint is to position the forecast reachable set of the pursuer as to maximize the probability of interception. The reachable set is restricted by the presence of hard actuation constraints. In a simulated engagement with noise corrupted measurements and non-Gaussian densities, the novel terminal constraint is shown to secure smaller miss distances as compared with predictive guidance based on the minimum mean square error or the maximum a-posteriori probability estimates