Fast fixed-time prescribed performance adaptive control for multi-aircraft formation reconfiguration based on BELNN...

Area of Science:

• Aerospace Engineering
• Control Systems
• Artificial Intelligence

Background:

• Formation reconfiguration in multi-aircraft systems (MAS) faces challenges in transient and steady-state performance.
• Existing control methods struggle with unknown disturbances and achieving rapid, precise convergence.

Purpose of the Study:

• To develop an advanced control strategy for enhancing MAS formation reconfiguration.
• To address transient overshoot, convergence speed, and steady-state errors under unknown disturbances.
• To improve formation security and control efficiency.

Main Methods:

• Proposed a Fast Fixed-Time Prescribed Performance Adaptive Control (FFTPAC) strategy.
• Integrated a Brain Emotional Learning Neural Network (BELNN) for disturbance estimation.
• Utilized a preset-time polynomial prescribed performance function for error bounds.
• Developed a fast variable power fixed-time reaching law for accelerated convergence.

Main Results:

• BELNN effectively estimates disturbances, reducing chattering and initial errors.
• The prescribed performance function ensures errors converge within a predefined time and range.
• The novel reaching law significantly increases system convergence rate.
• FFTPAC demonstrates superior control performance and robustness compared to existing methods.

Conclusions:

• The FFTPAC strategy achieves excellent control performance and robustness for MAS formation reconfiguration.
• The proposed method offers enhanced security and efficiency in dynamic flight operations.
• This work provides a significant advancement in adaptive control for complex multi-agent systems.