On Robust Control With Uncertainties in the Regressor Matrix and Parameter Vector

• 成果类型: Article
• 署名作者: Ebeigbe, Donald; Nguyen, Thang
• 署名单位: Pennsylvania Commonwealth System of Higher Education (PCSHE); Pennsylvania State University; Pennsylvania State University - University Park; Texas A&M University System; Thuyloi University
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2023.3264169
• 发表日期: 2024
• 页码: 309-314
• 关键词: Motion control; Passivity; regressor-based control; Robust control

摘要

Existing research results on regressor-based control use the linear parameterization of the dynamic equation into a known regressor matrix and an uncertain parameter vector. In this note, we show that the well-known robust controller that assumes a known regressor matrix can yield undesirable performance and poor robustness capabilities if there is uncertainty in the regressor matrix, especially when dealing with larger robots. The novelty of our result lies in accounting for uncertainties in both the regressor matrix and parameter vector, while ensuring that the uncertainty bound of our controller depends on a tuning parameter, which reflects the confidence in the regressor matrix accuracy, and the upper bound on the inertia parameters. Our controller, when compared against the controller that assumes a known regressor matrix, facilitates improved robustness while maintaining a smaller uniform ultimate boundedness radius.