Page 20 - ASME DSCC 2015 Program

Technical Program

experimental Verification of dynamic contour error estimation for design optimization of Solenoid actuated Butterfly Valves dynamically
High-Precision Contouring of Two-Axis Servo-Systems Coupled in Series
Contributed regular paper. DSCC2015-9744 Contributed regular paper. DSCC2015-9605
azad ghaffari, a. galip ulsoy, University of Michigan, Ann Arbor, MI, United Peiman naseradinmousavi, San Diego State University, San Diego, CA,
States United States, C. nataraj, Villanova University, Villanova, PA, United States
High-precision contouring is important in machining. A practically prov- In this effort, we present novel nonlinear modeling of two solenoid actuated
en method to improve contouring precision is adding a cross-coupling butterfly valves operating in series and then develop an optimal configura-
algorithm, which acts on contour error, to the existing position control tion in the presence of highly coupled nonlinear dynamics. The valves are
loops. The contour error estimate (CEE) significantly affects performance used in the so-called ‘smart Systems’ to be employed in a wide range of ap-
of the cross-coupling algorithm. Conventional CEE methods rely on static plications including bioengineering, medicine, and engineering fields. Typ-
single-point algorithms. A Newton-based CEE algorithm which effectively ically, tens of the actuated valves are instantaneously operating to regulate
improves CEE and dramatically reduces contouring error has recently been the amount of flow and also to avoid probable catastrophic disasters which
proposed. In this paper, instead of a separate cross-coupling control it is have been observed in the practice. We focus on minimizing the amount of
proposed to modify the position control loops, in this case integral sliding energy used in the system as one of the most critical design criteria to yield
mode control, to incorporate the contour error in the control loop. Various an efficient operation. We optimize the actuation subsystems interacting
experiments to identify the effect of 1) number of required iterations of the with the highly nonlinear flow loads in order to minimize a lumped amount
Newton-based CEE, 2) reference feedrate and curvature, and 3) sharp of energy consumed. The contribution of this work is to include coupled
corners on overall performance of the proposed cross-coupling algorithm nonlinearities of electromechanical valve systems to optimize the actuation
are reported. The experimental setup includes a two-axis servo-system. units. Stochastic, heuristic, and gradient based algorithms are utilized in
Control and estimation algorithms are implemented on two sbRIO 9632 from seeking the optimal design of two sets. The results indicate that substantial
National Instruments. amount of energy can be saved by an intelligent design that helps select pa-
rameters carefully but also uses flow torques to augment the closing efforts.
Adaptive Robust Cascade force Control of 1-Dof Joint Exoskeleton for
Human Performance Augmentation
ConTRIBuTED SESSIon
Contributed regular paper. DSCC2015-9825 1-4-1 Wa3 estimation and Identification 1
Shan chen, xiaocong Zhu, Shiqiang Zhu, Zhejiang University, Hangzhou, George Bellows E 10:00am–12:00pm
China, Bin Yao, Purdue University, West Lafayette, IN, United States, Zheng
Chen, Dalhousie University, Halifax, NS, Canada Session Chair: Jaspreet Dhupia, Nanyang Technological University
The control objective of exoskeleton for human performance augmentation Session Co-Chair: feitian Zhang, University of Maryland
is to minimize the human machine inter- action force while carrying external
Estimating Rigid Transformation With Correlated observations
loads and following human motion. This paper addresses the dynamics and
Contributed regular paper. DSCC2015-9672
the interaction force control of a 1-DOF hydraulically actuated joint exoskel-
eton. A spring with unknown stiffness is used to model the human- machine Chung-Yen lin, Masayoshi Tomizuka, University of California, Berkeley, CA,
interface. A cascade force control method is adopted with high-level con- United States
troller generating the reference position command while low level controller Estimating Rigid Transformation With Correlated Observations
doing motion tracking. Adaptive robust control(ARC) algorithm is developed observer Based Adaptive Estimation/Cancellation of unmatched
for both two controllers to deal with the effect of parametric uncertainties Sinusoidal Disturbances in Known lTI Systems by State Derivative
and uncertain nonlinearities of the system. The proposed adaptive robust Measurement
cascade force controller can achieve small human-machine interaction
Contributed regular paper. DSCC2015-9685
force and good robust performance to model uncertainty which have been
validated by experiment. Halil I. Basturk, Bogazici University, Istanbul, Turkey
In this paper, an adaptive observer and backstepping controller are designed
to cancel and estimate sinusoidal disturbances forcing a linear time-invariant
by using only the measurements of the state-derivatives. The parametrization
of the sinusoidal disturbance as the output of a known feedback system with
an unknown output vector that depends on unknown disturbance parame-
ters with the necessary filter designs enables to approach the problem as
an adaptive control problem. An observer is designed for the unmeasured
virtual input to apply a backstepping procedure which handles the unmatched
disturbance and input condition. Firstly, it is shown that the disturbance and
the unmeasured actuator state are observed perfectly in the open loop
case. Secondly, the closed loop case is considered and it is proven that the
equilibrium of the closed-loop adaptive system is stable and the state of the
considered original system converge to zero as t goes to infinity with perfect
20 disturbance estimation. The effectiveness of the controller and the observers
are illustrated with a simulation example of a third order system.