MPPI-Based Computational Guidance for Impact Time Control with Practical Constraints

Abstract

This paper presents a computational guidance algorithm to address the problem of impact time control with some practical constraints, such as the acceleration and seeker's field-of-view (FOV) limits. In this study, we adopt the Model Predictive Path Integral (MPPI) control method, which is an optimization approach utilizing the stochastic process to determine optimal solutions. The MPPI control can be considered as a data-driven method for solving nonlinear constrained optimization problems, and it has the potential for online implementation. With this mathematical tool, the proposed guidance algorithm is given in the form of proportional navigation (PN) guidance with a time-varying gain which is optimized at each guidance cycle by the iterative path integral in conjunction with the importance sampling under the model predictive control (MPC) setup. The proposed guidance does not rely on accurate time-to-go estimation and linearization assumptions compared to the existing methods. Since a negative gain value can also be exploited as a feasible solution under the proposed method, the achievable range of desired impact time can also be extended. Furthermore, unlike other computational guidance approaches, the proposed method can determine optimal solutions satisfying various practical constraints without requiring any dedicated solvers for optimization problems. Finally, we perform numerical simulations to validate the effectiveness of the proposed guidance strategy and compare it with other existing ones.