Cooperative control design. A systematic, passivity-based approach (Q545394): Difference between revisions

The book presents a passively-based framework that allows a systematic design of scalable and decentralized cooperative control laws. This framework makes explicit the passivity properties used implicitly in existing results and simplifies the design and analysis of a complex network of agents by exploiting the network structure and inherent passivity properties of agent dynamics. As it is demonstrated in the book, this passivity-based framework can be easily tailored to handle classes of cooperative control problems. Each of these problems has important applications in practice. For example, formation control and coordinated path following allow vehicles to maintain a tight relative configuration, thereby achieving effective group sensing or drag force reduction. Attitude coordination ensures that multiple spacecraft keep a precise relative attitude, which is important in space interferometry missions. Cooperative load transport enable robots to move an object around meanwhile exerting desired internal forces on the object. The intention of the book is to summarize in a coherent manner the authors' recent research in passivity-based cooperative control of multi-agent systems. Related passivity approaches are applicable to other interconnected systems, such as data communication networks, biomolecular systems, building Heating, Ventilation, and Air Conditioning (HVAC) systems, and power networks, though these applications are outside the scope of this book. The organization of the book is as follows: Chapter 1 introduces cooperative control and presents the necessary background in graph theory and passivity. Chapter 2 presents the passivity-based approach to cooperative control and applies it to the agreement problem. A coordination problem that is applicable to formation stabilization and group agreement as special cases is given and a class of feedback laws is presented that solve this problem with local information. The passivity approach also leads to a systematic construction of a Lyapunov function in the form of a sum of storage functions for the subsystems. Chapter 3 and 4 consider the situation where only one agent possesses the group reference velocity information and develop adaptive designs that recover such information for the other agents. Chapter 3 adopts an internal model approach while Chapter 4 parameterizes the reference velocity with time-varying basic functions. These two adaptive schemes illustrate the design flexibility offered by the passivity-based approach. Chapter 5 investigates attitude agreement of multiple rigid bodies. The results of Chapters 2 and 4 to a similar passivity-based framework that addresses the agreement of agents in SO(3) (Special Orthogonal group of dimension 3) are adapted. Chapter 6 studies agreement problem of multiple Euler-Lagrange systems. In particular, the case where uncertainty exists in agents' dynamics is considered. Two adaptive designs that ensure the agreement in the presence of uncertainty are presented. Chapter 7 presents a synchronized path following problem, where the agents achieve tracking of a desired formation by synchronizing path parameters. The synchronization of the path parameters is solved by the passivity-based agreement designs in Chapter 2. Chapter 8 studies cooperatively transporting a common load by multiple robots. This problem is formulated in a similar fashion to the position-based formation control in Chapter 2 and is validated with experiments. Chapter 9 provides an investigation of robustness properties for a second order linear cooperative control system. The Appendix includes technical proofs and tools utilized in the book.