Scholarlink

<p>This paper proposes an adaptive pitch control strategy that bridges state-dependent control and data-driven control. The method retains the SDRE feedback structure while reconstructing the required system matrices online from input-state data using sliding-window least-squares estimation, thereby eliminating the need for explicit analytical models. A Lyapunov-based boundedness analysis suggests uniformly ultimately bounded tracking behavior under bounded estimation errors and slowly varying parameters. Validation under step and sinusoidal references, measurement noise, and parameter mismatch demonstrates competitive tracking performance against analytical SDRE and PID controllers. A first-order low-pass filter applied to the identified matrices reduces control energy by up to 29% and total quadratic cost by 26% under noisy conditions. Sensitivity analyses for window length and the weighting matrices Q and R provide practical tuning guidelines. The simulation runs approximately four times faster than real time under the tested conditions, indicating potential for real-time implementation. The study illustrates how SDRE can be implemented in a data-driven manner while retaining its structural interpretability.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Adaptive nonlinear aircraft pitch control via LS-SDRE

  • 期刊论文

  • 发表日期：

    2026年6月27日

查看全文
  • Tung Le

摘要

This paper proposes an adaptive pitch control strategy that bridges state-dependent control and data-driven control. The method retains the SDRE feedback structure while reconstructing the required system matrices online from input-state data using sliding-window least-squares estimation, thereby eliminating the need for explicit analytical models. A Lyapunov-based boundedness analysis suggests uniformly ultimately bounded tracking behavior under bounded estimation errors and slowly varying parameters. Validation under step and sinusoidal references, measurement noise, and parameter mismatch demonstrates competitive tracking performance against analytical SDRE and PID controllers. A first-order low-pass filter applied to the identified matrices reduces control energy by up to 29% and total quadratic cost by 26% under noisy conditions. Sensitivity analyses for window length and the weighting matrices Q and R provide practical tuning guidelines. The simulation runs approximately four times faster than real time under the tested conditions, indicating potential for real-time implementation. The study illustrates how SDRE can be implemented in a data-driven manner while retaining its structural interpretability.