A Predefined-Time Optimization Approach for Distributed Resource Allocation of Multiagent Systems

This paper studies the distributed resource allocation problem of multiagent systems, which minimizes the team objective function while satisfying resource equality constraints. To achieve fast convergence, a predefined-time optimization approach is proposed based on distributed Laplacian-gradient algorithms and a sign-exp-power method. With the aid of predefined-time Lyapunov theory and convex optimization theory, it is proved that the proposed approach is convergent to the optimal solution within the predefined settling time, which can be adjusted through a controller parameter. Finally, numerical simulations confirm the effectiveness of the proposed optimization approach.