Resilient Multi-Dimensional Formation Control of MASs against Agent-Based and Denial-of-Service Attacks

In this paper, we consider the resilient multidimensional formation control of multi-agent systems (MASs) in the presence of both agent-based and denialof-service (DoS) attacks. The considered agent-based attacks encompass malicious, Byzantine, and stubborn behaviors, while DoS attacks can intermittently block the communication links among agents, causing the underlying digraph to become time-varying and even disconnected. The control objective is to guarantee that all benign agents achieve the desired formation despite these adversarial conditions. To this end, an "auxiliary point" based resilient control algorithm is proposed for MASs. Under the proposed algorithm, each healthy agent first performs a coordinate transformation into the formation error space and then constructs a "safe kernel" utilizing the transformed states of its in-neighbors. At each iteration, the agent updates its transformed state toward a specific point within this kernel, which guarantees resilient formation convergence. When the reception of neighbor states is disrupted by DoS attacks, the agent instead utilizes the most recently received information prior to the disruption. The effectiveness of the proposed approach is validated through theoretical analysis, numerical simulations, and hardware experiments on the Crazyflie quadrotor platform.