| 引用本文: | 王敏,胡锐,辛学刚,时昊天.严格反馈系统的事件触发学习控制[J].控制理论与应用,2021,38(10):1577~1586.[点击复制] |
| Wang Min,HU Rui,XIN Xue-gang,SHI Hao-tian.Event-triggered learning control for strict-feedback systems[J].Control Theory and Technology,2021,38(10):1577~1586.[点击复制] |
|
| 严格反馈系统的事件触发学习控制 |
| Event-triggered learning control for strict-feedback systems |
| 摘要点击 1962 全文点击 703 投稿时间：2020-10-20 修订日期：2021-09-17 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 10.7641/CTA.2021.00730 |
| 2021,38(10):1577-1586 |
| 中文关键词 确定学习 事件触发机制 自适应控制 神经网络 严格反馈系统 网络控制系统 |
| 英文关键词 deterministic learning event-triggered mechanism adaptive control neural networks strict-feedback systems network control system |
| 基金项目 国家自然科学基金项目(61773169, 61973129), 广东省自然科学基金项目(2019B151502058), 广东省重点领域研发计划项目(2020B1111010002), 广东海洋经济发展专项(粤自然资合[2020]018号), 广州市科技计划项目(201904010295)资助. |
|
| 中文摘要 |
| 本文针对一类严格反馈非线性系统, 提出了基于确定学习的事件触发控制方案. 首先, 在本地控制测试端
设计自适应神经网络控制, 并在控制过程中实现系统未知动态的知识获取和存储. 随后, 基于常值权值, 设计了新颖
的事件触发控制器和事件触发条件. 结合李雅普诺夫稳定性分析和非线性脉冲动态系统原理, 验证了所提方案能够
保证跟踪误差收敛到零的小邻域内以及所有闭环信号是最终一致有界的. 此外, 本文所提方案采用常值权值代替了
估计权值, 使得所提方案易于实现, 暂态性能好和网络资源占用少. 最后, 通过对比仿真结果证明了所提方案的有效
性. |
| 英文摘要 |
| This paper proposes a novel event-triggered tracking control scheme for a class of strict-feedback nonlinear
systems based on deterministic learning. Firstly, this paper designs an adaptive neural network control for strict-feedback
systems on the local control test side and realizes the knowledge acquirement of unknown dynamics in the control process.
Then, based on stored constant weights, this paper designs a novel event-triggered controller and a triggering condition. By
combining Lyapunov stability analysis with nonlinear impulse dynamic system theory, the proposed control scheme can
be verified to guarantee that the tracking error converges to a small neighborhood of zero and all the signals in the closedloop
system are uniformly ultimately bounded. Moreover, the proposed control scheme uses constant weights instead
of estimated weights, which makes the proposed scheme possess some good features including the easy-to-implement
triggered condition, the improved transient control performance, and the less network resource occupancy. A comparative
simulation is given to illustrate the effectiveness of the proposed scheme. |
|
|
|
|
|