|
NON-LINEAR KALMAN FILTER ESTIMATION FOR ACTIVE STEERING OF PROFILED RAIL-WHEELS
H. Li1, T.X. Mei2, J.T. Pearson1, and R.M. Goodall1
1 Department of Electronic and Electrical Engineering
Loughborough University, Loughborough, LEICS. LE11 3TU, UK
Email: J.T.Pearson@lboro.ac.uk, R.M.Goodall@lboro.ac.uk
2 School of Electronic and Electrical Engineering, University of Leeds,
Leeds, LS2 9JT, UK, Email:-t.x.mei@ee.leeds.ac.uk
This paper presents a study of state estimation for active steering of profiled wheels. Fundamental characteristics of a solid-axle wheelset are explained and the potential benefits of active steering are discussed. The paper studies the non-linearity of the wheelset caused by profiled wheel-rail contact surfaces and develops a recursive non-linear Kalman filter to provide reliable and accurate estimations of the wheelset state. Real wheel and rail profiles are used in the study and a non-linear model is developed to represent the wheelset. Computer simulations are used to verify the design and assess its performance.
Keywords: Active Control, State estimation, Railway wheelset, Wheel-rail profiles, Kalman filter, Non-linear system
Session slot T-Th-A01: Identification of Nonlinear Systems II/Area code 3a : Modelling, Identification and Signal Processing
 | 