RNN Learning-Based Prescribed-Time Safe and Robust Cooperative Group Formation Control for High-Speed Flight Vehicle...

Area of Science:

• Robotics and Control Systems
• Aerospace Engineering
• Artificial Intelligence

Background:

• Cooperative control of high-speed flight vehicles (HSFVs) faces limitations with complex, multi-target missions.
• Existing methods struggle with uncertainties like aerodynamic disturbances, actuator faults, and collision risks.

Purpose of the Study:

• To develop a safe and robust cooperative group formation control protocol for HSFVs under dynamic event-triggered communication.
• To address challenges posed by multiple compounding factors including uncertainties and potential intervehicle collisions.

Main Methods:

• Decomposition of HSFV fleet into subgroups for coordinated control.
• Development of a distributed prescribed-time event-triggered estimator (DP-TE-TE) for information acquisition.
• Application of constraint-following theory to manage collision and trajectory tracking constraints.
• Integration of a recurrent neural network (RNN) for online learning and compensation of unknown nonlinear dynamics.
• Proposal of a prescribed-time safe and robust cooperative group formation control scheme (P-TSRCGFCS).

Main Results:

• The DP-TE-TE successfully enables HSFVs to acquire necessary information for formation.
• Constraint-following theory effectively converts safety and tracking requirements into manageable constraints.
• The RNN provides online compensation for complex system uncertainties, enhancing control performance.
• Simulations demonstrate the effectiveness of the P-TSRCGFCS in achieving cooperative group formation for 12 HSFVs.

Conclusions:

• The proposed P-TSRCGFCS effectively manages complex aerial missions for multiple HSFVs.
• The integration of RNNs and event-triggered communication ensures safety and robustness in cooperative formations.
• The study provides a viable solution for advanced multi-vehicle aerial mission control.