Additive State Decomposition-Based Control and Guidance Subject to Singularity Phenomenon

• Published in: IEEE transactions on aerospace and electronic systems Vol. 59; no. 6; pp. 9067 - 9079
• Main Authors: Zhang, Xilian, Xu, Tao, Duan, Zhisheng, Yang, Jianying
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.12.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Additive state decomposition (ASD); Additives; Control systems; Decomposition; Feedback linearization; Guidance (motion); High gain; input saturation; Lead; Maneuvering targets; Missiles; nonlinear guidance law; Nonlinear systems; Singularities; singularity; Sliding mode control; State observers; Task analysis; Tracking problem; Uncertainty; unknown disturbance
• ISSN: 0018-9251; 1557-9603
• DOI: 10.1109/TAES.2023.3314712

In this article, an additive state decomposition (ASD)-based method is proposed to achieve control and guidance tasks subject to singularity phenomenon, where input saturation and unknown disturbance are taken into consideration. The singularity phenomenon considered in this article refers to that the control gain in the denominator of the feedback-linearization methods approaches zero, resulting in infinitely large control amplitudes. Based on the concept of ASD, a novel ASD-based controller is proposed where the control system is decomposed into a primary system and a secondary system. Specifically, a nonsingular nonlinear system with unknown disturbance is included in the primary system and a nonlinear system with input saturation is included in the secondary system. Accordingly, the tracking problem for the original control system is decomposed into the extended high-gain state observer-based tracking problem for the primary system and the stabilization problem for the secondary system. In addition, an ASD-based all-aspect guidance law against maneuvering target with unknown target acceleration and input saturation is developed, which enables the missiles to intercept maneuvering targets from any initial heading angles. Finally, the effectiveness and superiority of the proposed methods are demonstrated by some numerical simulation examples.