Meta-heuristic optimization-based robust H∞ controller design for active suspension systems subject to actuator saturation

• Published in: Alexandria engineering journal Vol. 105; pp. 523 - 537
• Main Authors: Kim, Yeongjae, Kim, Myoung-Woo, Kanno, Masaaki, Kim, Tae-Hyoung
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.10.2024; Elsevier
• Subjects: Active suspension systems; Actuator saturation; Bilinear matrix inequality; H∞control; Meta-heuristic algorithm; Output feedback; State feedback; State feedback; Meta-heuristic algorithm; H∞control; Active suspension systems; Output feedback; Actuator saturation; Bilinear matrix inequality
• ISSN: 1110-0168
• DOI: 10.1016/j.aej.2024.08.032

This study focuses on designing robust H∞ controllers to mitigate actuator saturation-related issues of active suspension systems. Two design methods are presented: one utilizing full-state feedback and the other relying on measurable output feedback. The H∞ state-feedback control incorporates actuator saturation directly into the design, representing the control input with the saturation function in the state-space model. The resulting optimization-based design problem, derived from theoretical stability analyses, poses a non-convex NP-hard problem under bilinear matrix inequality (BMI) constraints. To address this challenge, the study proposes a meta-heuristic optimization-based technique, providing a straightforward yet effective method to manage the BMI problem. The scope extends to practical implementation considerations, where suspension systems may only utilize measurable output variables. The H∞ static output-feedback control case is explored by extending the design of a full-order state-feedback controller. Despite the BMI-related challenges entailed in this scenario, the proposed meta-heuristic optimization-based design technique demonstrates flexibility in managing BMI conditions, thus eliminating the need to modify the controller design methodology. A noteworthy contribution of this study is its consistent application of the same meta-heuristic optimization-based technique for designing both H∞ state-feedback and static output-feedback controllers. Experimental results validate the effectiveness of the proposed robust H∞ control strategies.