Convex Temporal Convolutional Network-Based Distributed Cooperative Learning Control for Multiagent Systems.

Area of Science:

• Control Systems Engineering
• Artificial Intelligence
• Nonlinear Dynamics

Background:

• Distributed cooperative learning control is efficient and scalable but faces challenges in modeling uncertainties in multiagent systems.
• Exploiting cooperative learning in complex, uncertain systems requires advanced modeling techniques.

Purpose of the Study:

• To propose a novel distributed cooperative learning control method for uncertain discrete-time nonlinear multiagent systems.
• To address the challenges of modeling uncertainties and leveraging cooperative learning in such systems.

Main Methods:

• A convex temporal convolutional network (CTCNet) is proposed for cooperative estimation of uncertain agent dynamics.
• Learned knowledge is shared through communication topology among CTCNet adaptive laws to improve control performance.
• The method is designed for robustness across different control tasks without requiring network weight adjustments.

Main Results:

• The proposed CTCNet-based distributed cooperative learning control method demonstrates improved control accuracy.
• The method exhibits enhanced robustness and anti-interference capabilities compared to existing approaches.
• Asymptotic convergence of system tracking errors to the origin is strictly proven.

Conclusions:

• The novel CTCNet-based approach effectively handles uncertainties in discrete-time nonlinear multiagent systems.
• This method offers superior performance, robustness, and anti-interference ability.
• The findings validate the potential of convex temporal convolutional networks in advanced control applications.