Distribution consensus of multi-Agent systems based on model predictive control with probability density function

A distributed output feedback model predictive control approach based on the probability density function method is proposed to realize multi-Agent distribution consensus. Each agent solves the optimal control input by estimating the worst-case local error and perturbation, modeled into a local min-max optimization problem. In the iterative solving process, the agent i will send its information to its neighbor agent through the communication topology, so as to achieve the convergence of group consensus error. Under the assumption of controllability and observability, the proposed control method provides an upper bound for the group distribution consensus error, thus ensuring the practical distribution consensus performance under unmeasured interference and noise.