000 05679nas|a2200397 a 4500
005 20260817171017.0
008 100807s2008 th uu|m rtt 0| a1eng d
035 _a.b12081346
099 9 _aAIT Diss. no.ISE-08-04
100 0 _aBui Trung Thanh
245 1 0 _aBalancing control of bicycle robot by particle swarm optimization-based structure-specified H₂/Hoo control
260 _aPathum Thani, Thailand :
_bAsian Institute of Technology,
_c2008
300 _a1 online resource (88 p.) :
_bill.
490 1 _aDissertation ;
_vno. ISE-08-04
500 _aA dissertation submitted in partial fulfillment of the requirements for the Degree of Doctor of Engineering in Mechatronics, School of Engineering and Technology
502 _aThesis (Ph.D.) - Asian Institute of Technology, 2008
520 _aThe purpose of this dissertation is to develop a prototype of bicycle robot, Bicyrobo, and then propose advanced robust and optimal control algorithms to control balancing of the system. This dissertation proposed two novel H₂/Hoo control algorithms: (i) Particle swarm optimization (PSO)-based structure-specified Hoo loop shaping control, and (ii) Particle swarm optimization-based structure-specified mixed H₂/Hoo control. Firstly, a prototype of bicycle robot with a flywheel system, Bicyrobo, is designed and developed. This part of the dissertation describes the design and manufacturing of mechanical parts and the building of electrical parts, including an embedded PC as a center controller and all signal conditioning circuits for interfacing to sensors and motors. A simplified dynamics model of the system is then derived based on Lagrange method. Stability analysis shows that the system is unstable without any control effort. In the next part of the dissertation, a PSO-based structure-specified Hoo loop shaping controller design algorithm is proposed. The algorithm is developed based on the Hoo loop shaping control technique initially proposed by McFarlane and Glover. In the proposed method, model uncertainties of the system are represented by normalized coprime factor uncertainties. The nominal model of the system is firstly shaped by a pre-compensator and a post-compensator to achieve the desired open loop shape. A structure-specified controller is then defined. Finally, PSO algorithm is used to search for parameters of this structure-specified controller so that the controller is admissible and the Hoo norm from exogenous inputs to controlled outputs, which follows the standard setup of Hoo loop shaping control theory, is minimized. The proposed algorithm is successfully applied to design the first and second order Hoo loop shaping controllers to control balancing of Bicyrobo. Which is the unstable system involved with sources of uncertainties due to un-modeled dynamics, parameter variations and external disturbances. Simulation and experimental results show the robustness and efficiency of the proposed controllers in compared with the proportional plus derivative (PD) controller, and the conventional Hoo loop shaping controller. The proposed PSO-based algorithm also shows superior to the GA-based algorithm for this optimization problem. Another control algorithm, PSO-based structure-specified mixed H₂/Hoo controller design algorithm, is also proposed in this dissertation. The structure-specified mixed H₂/Hoo controller design aims to design a low order controller to attain both robust stability and good performance, for instance, small tracking error, less control energy, etc. In the proposed method, model uncertainties of the system are represented by multiplicative uncertainties, and the system is also assumed to be affected by external disturbances. A structure-specified controller is firstly defined. Then, the PSO algorithm is applied to search for parameters of an admissible structure-specified controller that minimizes an integral of squared error (H₂ norm) subjected to robust stability constraints (Hoo norm) against both model uncertainties and external disturbances. The proposed algorithm is successfully applied to design a first order mixed H₂/Hoo controller to control balancing of Bicyrobo. Simulation and experimental results on the system verify the robustness of the proposed controller in compared with the conventional proportional plus derivative (PD) controller. The simulation results also verify the efficiency of the PSO-based algorithm in compared with the GA-based algorithm in term of computational effort, computational time, and convergent speed
650 0 _aBicycles
_xDynamics
700 0 _aManukid Parnichkun,
_eChairperson
700 1 _aAfzulpurkar, Nitin V.,
_eExamination Committee
700 0 _aWeerakorn Ongsakul,
_eExamination Committee
710 2 _aMinistry of Education and Training, Vietnam,
_eScholarship Donor
710 2 _aAsian Institute of Technology Fellowship,
_eScholarship Donor
810 2 _aAsian Institute of Technology.
_tDissertation ;
_vno. ISE-08-04
856 _3Full-Text
_uhttp://203.159.5.9/ait-thesis/detail.php?q=B00649
907 _a.b12081346
_bmnait
_cu
902 _a240329
998 _b0
_c100807
_dm
_ea
_fu
_g0
945 _lmnait
945 _lmnarc
942 _c22
942 _c40
909 _aBarcode : 30050120573729
_bCREATED : 2012-08-31
_cRECORD # : i12692037
_dLPATRON : 1027674
_eLCHKIN : 2018-07-24
_f# RENEWALS : 0
_g# OVERDUE : 0
_hIUSE3 : 6
_iTOT CHKOUT : 5
_jTOT RENEW : 4
909 _aBarcode : 30050160002290
_bCREATED : 2016-02-17
_cRECORD # : i12883918
_dLPATRON : 0
_eLCHKIN : -
_f# RENEWALS : 0
_g# OVERDUE : 0
_hIUSE3 : 0
_iTOT CHKOUT : 0
_jTOT RENEW : 0
999 _c10553
_d10553