This paper concerns the disturbance rejection problem of a linear complex dynamical network subject to external disturbances. A dynamical network is said to be robust to disturbance, if the H∞ norm of its transfer function matrix from the disturbance to the performance variable is satisfactorily small. It is shown that the disturbance rejection problem of a dynamical network can be solved by analysing the H∞ control problem of a set of independent systems whose dimensions are equal to that of a single node. A counter-intuitive result is that the disturbance rejection level of the whole network with a diffusive coupling will never be better than that of an isolated node. To improve this, local feedback injections are applied to a small fraction of the nodes in the network. Some criteria for possible performance improvement are derived in terms of linear matrix inequalities. It is further demonstrated via a simulation example that one can indeed improve the disturbance rejection level of the network by pinning the nodes with higher degrees than pinning those with lower degrees.
This paper presents a coordinating and stabilizing control law for a group of underwater vehicles with unstable dynamics. The coordinating law is derived from a potential that only depends on the relative configuration of the underwater vehicles. Being coordinated,the group behaves like one mechanical system with symmetry,and we focus on stabilizing a family of coordinated motions,called relative equilibria. The stabilizing law is derived using energy shaping to stabilize the relative equilibria which involve each vehicle translating along its longest（unstable） axis without spinning,while maintaining a relative configuration within the group. The proposed control law is physically motivated and avoids the linearization or cancellation of nonlinearities.
