Differentially Private Synthesis for Cooperative Output Regulation of Discrete-Time Linear Multi-Agent Systems

This work investigates the differentially private algorithm for the cooperative output regulation problem when the reference signal is constant. A distributed control strategy is developed that ensures individual privacy by adding random noises to the exchanged information. Moreover, a set of time-varying gains is formulated to ease the condition that the imposed noise must decrease exponentially to zero. Furthermore, a method is proposed to design the control gains to ensure that multi-agent systems maintain a specified differential privacy index. Lastly, a numerical example is presented to demonstrate the practical effectiveness of the proposed control strategy.