0433945 20240111140853.020141030235959.9 84916931171 000400962700154 10.1109/MED.2014.6961492 6 s. C cav_un_epca*0433944978-1-4799-5899-3924-929PalermoIEEE2014 Identification Walking robots Regression cav_un_auth*0279330 Teorie řízení Department of Control Theory TŘ TR Department of Control Theory Dolinský Kamil UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0101074 Teorie řízení Department of Control Theory TŘ TR Department of Control Theory Čelikovský Sergej UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. textový dokument 323 kB cav_un_auth*0283207 GAP103/12/1794 GA ČR CZ Abstract—In this article we present a novel approach to identification of a class of mechanical systems that can be written in a form of Euler-Lagrage equations. This identification procedure is based on a polynomial regression that is employed for data smoothing and estimation. Procedure requires only the measurements of torques or forces that are used to control the system and measurements of angles or other co-ordinates that are used to specify the location of the system. This identification algorithm is demonstrated on the problem of identification of three link bipedal robot and the results indicate that the procedure is effective and over-performs classical approaches used to solve the problem. cav_un_auth*0308759 The 2014 22nd Mediterranean Conference on Control and Automation (MED) 16.06.2014-19.06.2014 Palermo IT BC 2015 2 PR RVO:67985556 http://hdl.handle.net/11104/0238108 cav_un_auth*0298829 České vysoké učení technické v Praze, Fakulta elektrotechnická CZ S 2014 84916931171 SCOPUS 000400962700154 WOS textový dokument 323 kB cav_un_epca*0433944 Proceedings of the 2014 22nd Mediterranean Conference on Control and Automation (MED) 978-1-4799-5899-3 924 929 Palermo IEEE 2014