Resilient Distributed Cooperative Backstepping Control of Euler–Lagrangian Systems Under Malicious Link Attacks

The resilient cooperative control issue for uncertain multi-Euler-Lagrangian (EL) systems under a locally-finite number of adversaries that may arbitrarily manipulate communication links is investigated in this paper. We propose a resilient distributed backstepping control strategy and characterize the corresponding necessary and sufficient condition to guarantee the asymptotic consensus to the target state regardless of arbitrarily injected malicious communication information. To restrain the impacts of adversaries, the filtered coordinate-based weighted Mean-Subsequence-Reduced (CW-MSR) mechanism is adopted, which induces the unknown topology switching and may cause the discontinuity of the virtual controls when applying the backstepping method. To overcome the difficulty, locally uniform adjacency weights are selected in the CW-MSR mechanism, and consequently the continuity of the (non-smooth) virtual controls is ensured even in the presence of the topology changes. The non-smooth analysis method is given to prove the asymptotic consensus for the considered system.