Coordinated Dynamics And Control Of Networked Euler-lagrange Systems

Coordinated dynamics and control of networked Euler-Lagrange (EL for short)systems has recently attracted a great deal of attention from various felds of scienceand engineering, which originates from two elementary reasons: Firstly, networkis regarded as a good tool to deal with the complex system compose of a set ofinteractional subsystems. Secondly, the networked EL systems plays an importantrole in various engineering applications involving complex and integrated produc-tion processes. A large variety of mechanical systems such as robotic manipulators,formation fying spacecrafts, autonomous vehicles, and multiple walking robots canbe well formulated by the networked EL systems. This dissertation is mainly con-cerned with the issues of synchronization motion and tracking control of networkedmechanical systems modeled by EL dynamics. In details, the fundamental contri-butions of this work is summarized in the following three aspects:1. Impulsive synchronization motion in networked EL systems. In general,impulsive motion occurs when the systems are physically subject to either director indirect impulsive efects. The impulsive efects are usually caused by impulsiveforces or impulsive constraints. By using preliminary feedback control and impul-sive constraints, the procedure for studying impulsive synchronization motion innetworked open-loop EL systems is presented. Some generic criteria on synchro-nization of the system output with respect to generalized coordinates and its veloc-ities are established respect to undirected fxed and switching network topologies.The procedure shows that impulsive synchronization motion in networked open-loop EL systems can achieve by impulsive constraints strategies. Some examplesand simulations show the efectiveness of the procedures.2. Backstepping tracking control in networked nonidentical EL systems. Thedistributed adaptive tracking control algorithms are designed for EL systems withuncertain external disturbance and parametric uncertainties. It is shown that theproposed algorithms only require a subset of the followers access to leader’s posi-tion, where the directed network topology graph has a spanning tree, and both theleader’s position and its velocity are assumed to be varying. Furthermore, the simu-lations show that the adaptive control algorithm for uncertain external disturbance systems is robust yet efciently eliminate the chattering during tracking.3. Distributed coordinated adaptive tracking in networked redundant roboticsystems. The distributed coordinated adaptive tracking algorithms of networkedredundant robotic systems with a dynamic leader are investigated. We providean analysis procedure for the controlled synchronization of such systems with un-certain dynamics, where the underlying directed graph has a spanning tree. Wealso fnd that the proposed control strategy does not require computing positionalinverse kinematics and does not impose any restriction on the self-motion of themanipulators, therefore, the extra degrees of freedom are applicable for other so-phisticated subtasks. Compared with some existing work, the distinctive featuresof the designed distributed control algorithm are that only a subset of followersneeds to access the position information of the dynamic leader in the task space.In addition, the viscous and dynamic friction of the systems are considered.