Transactions of Nanjing University of Aeronautics & Astronautics
Prescribed Performance Backstepping Control of Flexible Joint Based on Unknown State Estimator and Improved Tracking Differentiator
- Article
- Figures
- Metrics
- Preview PDF
- Reference
- Related
- Cited by
- Materials
Abstract:
In order to solve the problem of increasing tracking error of flexible joint system caused by model uncertainty and external unknown disturbance， a prescribed performance backstepping control method based on unknown state estimator and tracking differentiator is proposed. An unknown state estimator based on low-pass filter is designed， which can estimate the lumped disturbance only depending on the nominal value of the model. A novel finite-time convergence prescribed performance function is constructed. Based on this function， a backstepping controller is designed to ensure that all signals in the closed-loop system are bounded， and the joint tracking error converges to any given small area within the given time. To avoid the differential explosion of the backstepping controller， a tracking differentiator based on the improved Sigmoid function is designed to estimate the differential signal of virtual control law. The simulation results show that the proposed method can guarantee the joint tracking error converges to any given range within the finite time under the disturbance of unmodeled dynamics and unknown input， and can effectively improve the transient and steady-state performance of flexible joints.
Keywords:
Project Supported:
This work was supported by the National Natural Science Foundation of China （No.62076152）， the National Natural Science Foundation of China （No.62073200）， the Natural Science Foundation of Shandong Province （No.ZR2020MF096）， and funded by Beijing Advanced Innovation Center for Intelligent Robots and Systems.
Clc Number:
TP273
Mobile website
