Font Size: a A A

Adaptive robust force control for vehicle active suspensions

Posted on:2002-08-04Degree:Ph.DType:Dissertation
University:University of MichiganCandidate:Chantranuwathana, SupavutFull Text:PDF
GTID:1462390011990234Subject:Engineering
Abstract/Summary:
The objective of this dissertation is to develop high-performance force controllers for vehicle active suspensions. Since the majority of active suspension controllers in the literature did not include actuator dynamics, few of them have been verified to perform well experimentally. To provide accurate force generation, Adaptive Robust Control techniques (ARC) are adopted in this dissertation because they combine the advantages of deterministic robust control techniques and adaptive control techniques while avoiding many of their drawbacks. We also introduced a few enhancements to further improve the practicality and performance of the ARC algorithm. We first design a force controller by using existing ARC techniques. Secondly, an output feedback ARC is developed to make the resulting force controller easier to implement. An online adaptation algorithm is then proposed to improve performance and robustness of the controller. Finally, a modular technique to improve the controller's identification process is developed.; The importance of these enhancements is as follows. The output feedback extension is important for reducing sensor cost. The online adaptation algorithm improves system performance during normal operations while helping to stabilize the system under unmodeled uncertainties. The Modular ARC technique, which is the most important theoretical contribution of this work, improves the performance of the overall control system by allowing the identification algorithms to be designed with an emphasis on identification accuracy rather than on output tracking. Generalized procedures for designing an ARC controller with online adaptation of its performance parameters and a Modular ARC controller for systems in the parametric strict-feedback form are also provided.; The performance of the proposed ARC force control algorithms is verified experimentally in the University of Michigan Vehicle Dynamics Laboratory. Experimental results from a quarter-car test rig confirm that these force controllers work satisfactorily, either stand-alone or when they are integrated with LQ or LQG outer-loop control laws.
Keywords/Search Tags:Force, Controller, Active, Vehicle, ARC, Performance, Robust, Adaptive
Related items