A Learning-Based Computational Impact Time Guidance

TL;DR

This paper introduces a learning-based guidance algorithm that combines deep neural networks and reinforcement learning to accurately control impact time in missile guidance scenarios.

Contribution

It develops a novel prediction-correction guidance method integrating neural networks and reinforcement learning to improve impact time control accuracy.

Findings

The algorithm effectively estimates time-to-go with realistic aerodynamics.

Reinforcement learning addresses sparse reward issues in guidance.

Numerical simulations validate the approach's effectiveness.

Abstract

This paper investigates the problem of impact-time-control and proposes a learning-based computational guidance algorithm to solve this problem. The proposed guidance algorithm is developed based on a general prediction-correction concept: the exact time-to-go under proportional navigation guidance with realistic aerodynamic characteristics is estimated by a deep neural network and a biased command to nullify the impact time error is developed by utilizing the emerging reinforcement learning techniques. The deep neural network is augmented into the reinforcement learning block to resolve the issue of sparse reward that has been observed in typical reinforcement learning formulation. Extensive numerical simulations are conducted to support the proposed algorithm.