講座主題：魯棒模型預(yù)測(cè)控制和視覺伺服的設(shè)計(jì)框架及其在自主飛行器中的應(yīng)用
主講人：施陽(yáng)
講座時(shí)間：2022.6.17 上午9:30
鏈接：https://meeting.tencent.com/dm/yLsNKvEFkcfo
騰訊會(huì)議ID：475-402-785
專家介紹：加拿大維多利亞大學(xué)教授，1. Highly Cited Researcher, Clarivate Analytics (formerly Thomson Reuters), Listed for 8 years (2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021)
2. Fellow of Engineering Institute of Canada (EIC), December 2018 (加拿大工程研究院 院士)
3. Humboldt Research Fellowship for Experienced Researchers, December 2017
4. Fellow of IEEE, Jan. 2017
5. 2016 IEEE Tansactions on Fuzzy Systems Outstanding Paper Award, November 2016
6. Fellow of ASME, March 2016
7. Fellow of CSME, March 2016
8. Craigdarroch Silver Medal for Excellence in Research Award, University of Victoria, 2015
9. JSPS Invitation Fellowship (Short-Term), Japan Society for the Promotion of Science, 2013
10. Teaching Excellence Award, Faculty of Engineering, University of Victoria, April 2012
11. Leaders Opportunity Fund Award, Canada Foundation for Innovation, Dec. 2011
12. Teaching Excellence Award, University of Saskatchewan Student Union (USSU), April 2007"
報(bào)告內(nèi)容簡(jiǎn)介：Abstract: Model predictive control (MPC) is a promising paradigm for high-performance and cost-effective control of complex dynamic systems. This talk will introduce a unified framework of integrating robust model predictive control (MPC) and visual servoing for autonomous quadrotors. By combining the image kinematics with the quadrotor dynamics, we develop a robust NMPC based IBVS (Image-Based Visual Servoing) scheme to ensure the satisfaction of both physical and visibility constraints. The tightened state constraints are constructed for handing external disturbances. The sufficient conditions on guaranteeing recursive feasibility of the proposed MPC algorithm are established. Furthermore, we theoretically show that the closed-loop system is input-to-state practically stable (ISpS). Some existing challenges and future research directions will be discussed.
