Distributed finite-time optimisation algorithm for second-order multi-agent systems subject to mismatched disturbances

• Published in: IET control theory & applications Vol. 14; no. 18; pp. 2977 - 2988
• Main Authors: Wang, Xiangyu, Wang, Guodong
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 17.12.2020
• Subjects: control design; control system synthesis; distributed control; distributed finite‐time optimisation control algorithm; distributed finite‐time optimisation problem; feedback; feedforward; feedforward–feedback composite control framework; finite‐time consensus; finite‐time disturbance observer; finite‐time estimator techniques; finite‐time optimisation algorithm; homogeneous systems theory; integral sliding‐mode control; mismatched disturbances; multi‐agent systems; multi‐robot systems; observers; optimisation; Research Article; second‐order multiagent systems subject; second‐order single system; variable structure systems; distributed finite-time optimisation control algorithm; distributed finite-time optimisation problem; multi-agent systems; second-order multiagent systems subject; distributed control; finite-time consensus; multi-robot systems; control system synthesis; finite-time optimisation algorithm; finite-time estimator techniques; finite-time disturbance observer; feedback; integral sliding-mode control; feedforward–feedback composite control framework; optimisation; feedforward; mismatched disturbances; control design; observers; variable structure systems; homogeneous systems theory; second-order single system
• ISSN: 1751-8644; 1751-8652
• DOI: 10.1049/iet-cta.2020.0901

In this study, the distributed finite-time optimisation problem is investigated for second-order multi-agent systems subject to mismatched disturbances. The whole control design is conducted within a feedforward–feedback composite control framework, which consists of two main sessions. In the first session, based on homogeneous systems theory, a finite-time optimisation control algorithm is presented for a nominal second-order single system with a quadratic cost function. In the second session, via the result in the first session and by combining finite-time disturbance observer, homogeneous systems theory, integral sliding-mode control, and distributed finite-time estimator techniques together, a distributed finite-time optimisation control algorithm is proposed for second-order multi-agent systems with mismatched disturbances and quadratic local cost functions. Under the proposed control algorithm, the outputs of all the agents reach finite-time consensus on the minimiser of the global cost function. Simulations demonstrate the effectiveness and superiority of the proposed distributed finite-time optimisation control algorithm.