Reduction in the Amount of Data for Data-driven Passivity Estimation

The estimation of passivity based on input–output data without a system model is discussed. In the conventional data-driven estimation method based on the gradient approach, the convergence is slow and a large number of experiments is required. Therefore, we propose a method to improve the convergence, thereby reducing the amount of data required for estimation.

[1]  Panos J. Antsaklis,et al.  On relationships among passivity, positive realness, and dissipativity in linear systems , 2014, Autom..

[2]  Frank Allgöwer,et al.  Sampling strategies for data-driven inference of passivity properties , 2017, 2017 IEEE 56th Annual Conference on Decision and Control (CDC).

[3]  Frank Allgöwer,et al.  One-Shot Verification of Dissipativity Properties From Input–Output Data , 2019, IEEE Control Systems Letters.

[4]  Masaya Tanemura,et al.  Closed-Loop Data-Driven Estimation on Passivity Property , 2019, 2019 IEEE Conference on Control Technology and Applications (CCTA).

[5]  David J. Hill,et al.  Passivity-based control and synchronization of general complex dynamical networks , 2009, Autom..

[6]  Blake Hannaford,et al.  Time domain passivity control of haptic interfaces , 2001, Proceedings 2001 ICRA. IEEE International Conference on Robotics and Automation (Cat. No.01CH37164).

[7]  Frank Allgöwer,et al.  Some Ideas on Sampling Strategies for Data-Driven Inference of Passivity Properties for MIMO Systems , 2018, 2018 Annual American Control Conference (ACC).

[8]  Bernd Henze,et al.  Passivity-based whole-body balancing for torque-controlled humanoid robots in multi-contact scenarios , 2016, Int. J. Robotics Res..

[9]  John T. Wen,et al.  Building temperature control: A passivity-based approach , 2012, 2012 IEEE 51st IEEE Conference on Decision and Control (CDC).

[10]  Shun-ichi Azuma,et al.  Efficient Data-Driven Estimation of Passivity Properties , 2019, IEEE Control Systems Letters.