Generalized and Heterogeneous Nonlinear Dynamic Multiagent Systems Using Online RNN-Based Finite-Time Formation Tracking Control and Application to Transportation Systems

In this article, an online RNN-based finite-time formation tracking control (ORNN-FTFTC) is designed to quickly accomplish an assigned formation of nonlinear generalized and heterogeneous multiagent systems with input fault and saturation. Each agent, including the leader and the followers, can possess different relative degrees and control input numbers but the same output for easy task planning. At least one agent must communicate with the leader and the information of neighborhood agents is required to accomplish the assigned formation task. To fulfill the task under the uncertain environment, the proposed ORNN-FTFTC possesses nonlinear filtering formation error with dynamic fractional exponent, nonlinear filtering gain, and the RNN learning compensation of the aggregately uncertain dynamics in each agent. Not only does the nonlinear filtering gain increase as the nonlinear filtering formation error is in the vicinity of zero to achieve its finite-time convergence, but also the new e-modification learning can cover all the value of formation error such that learning weights are stabilized even in an uncertain environment. Finally, an application to a 3D pose from take-off to a steady-formation of hexa-copter unmanned aerial vehicles and unmanned helicopters with initial formation error certifies the feasibility and robustness of the proposed control.