- author = "Elias Jeremias Virgil Reichensdeorfer",
- title = "Traction Control via Input-Output Linearization and Stability Analysis using Parametric Lyapunov Functions",
- school = "Fakultaet fuer Elektrotechnik und Informationstechnik, Technische Universitaet Muenchen",
- year = "2022",
- address = "Munich, Germany",
- month = "23 " # may,
- note = "with distinction",
- keywords = "genetic algorithms, genetic programming, Traction Control, Input-Output Linearization, Zero Dynamics, Lyapunov Stability, Lur'e System, BMW Bayerische Motoren Werke, VDC Vehicle Dynamics Control",
- language = "en",
-
URL = "
https://www.ce.cit.tum.de/en/lsr/news/article/doktortitel-mit-auszeichnung-fuer-herrn-elias-reichensdoerfer/",
- broken = "https://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:91-diss-20220523-1618422-1-8",
-
URL = "
https://mediatum.ub.tum.de/doc/1618422/1618422.pdf",
- size = "164 pages",
- abstract = "The development of automotive vehicles over the past decades has led to engines with ever increasing power and faster dynamics. This development sets new opportunities for electronic traction control systems as new control designs are required to take advantage of the full potential of modern engines. This work proposes an approach to traction control for automotive vehicles with rear-wheel drive, front-wheel drive or on-demand four-wheel drive based on the method of input-output linearization. Three different control design models are analysed, that apart from the actuator dynamics, the wheel dynamics and the longitudinal vehicle dynamics, take the torsional dynamics of the drive train into account explicitly. Global asymptotic stability of the resulting zero dynamics is shown analytically for the control laws based on these design models. Stability results are derived for the whole class of each design model by using parametric Lyapunov functions. A reformulation of the zero dynamics of the rear-wheel drive design model as a Lure system is proposed. It is shown how passivity based methods can be combined with the proposed class of Lyapunov functions to strengthen the results to global exponential stability and input-to-state stability. A heuristic method for Lyapunov function identification based on genetic programming is proposed and its performance is evaluated on three nonlinear example systems.The static control laws for input-output linearisation are approximated by dynamic control laws for a robust implementation on the relevant control units. An experimental evaluation with different test vehicles is carried out and a comparison to traditional traction control systems is given. It is shown that the proposed traction control systems achieve better tracking performance,disturbance attenuation and damping of drive train oscillations.",
-
notes = "in English
Supervisor: Dirk Wollherr",
