A survey of models, analysis tools and compensation methods for the control of machines with friction

Past, present and future of nonlinear system identification in structural dynamics

1. Introduction and Historical Background 1.1 Why make human figure models? 1.2 Historical roots 1.3 What is currently possible? 1.4 Manipulation, animation, and simulation 1.5 What did we leave out? 2. Mody Modeling 2.1 Geometric body modeling 2.2 Representing articulated figures 2.3 A flexible torso model 2.4 Shoulder complex 2.5 Clothing models 2.6 The anthropometry database 2.7 The anthropometry spreadsheet 2.8 Strength and torque display 3. Spatial Interaction 3.1 Direct manipulation 3.2 Manipulation with constraints 3.3 Inverse kinematic positioning 4. Behavioral Control of Articulated Figures 4.1 An interactive system for postural control 4.2 Interactive manipulation with behaviors 4.3 The animation interface 4.4 Human figure motions 4.5 Virtual human control 5. Simulation with Societies of Behaviors 5.1 Forward simulation with behaviors 5.2 Locomotion 5.3 Strength guided motion 5.4 Collision-free path and motion planning 5.5 Posture planning 6. Task-level Specifications 6.1 Performance simulation with simple commands 6.2 Natural language expressions of kinematics and space 6.3 Task-level simulation 6.4 A framework for instruction understanding 7. Epilogue 7.1 A road map toward the future 7.2 Conclusion

/pdf/simulating-humans-computer-graphics-animation-and-control-2dk1fy50vu.pdf

Simulating humans: computer graphics animation and control

Two noniterative subspace-based algorithms which identify linear, time-invariant MIMO (multi-input/multioutput) systems from frequency response data are presented. The algorithms are related to the recent time-domain subspace identification techniques. The first algorithm uses equidistantly, in frequency, spaced data and is strongly consistent under weak noise assumptions. The second algorithm uses arbitrary frequency spacing and is strongly consistent under more restrictive noise assumptions, promising results are obtained when the algorithms are applied to real frequency data originating from a large flexible structure.

/pdf/subspace-based-multivariable-system-identification-from-18eqaqh7jc.pdf

Subspace-based multivariable system identification from frequency response data

This paper gives a survey of frequency domain identification methods for rational transfer functions in the Laplace (s) or z-domain. The interrelations between the different approaches are highlighted through a study of the (equivalent) cost functions. The properties of the various estimators are discussed and illustrated by several examples. >

Parametric identification of transfer functions in the frequency domain-a survey

The analysis of the periodic components in noise and vibration signals measured on rotating equipment such as car power trains, must be done more and more under rapid changes of an axle, or reference RPM. Normal tracking filters (analog or digital implementations) have limited resolution in such situations; wavelet methods, even when applied after resampling the data to be proportional to an axle RPM, must compromise between time and frequency resolution. The authors propose the application of nonstationary Kalman filters for the tracking of periodic components in such noise and vibration signals. These filters are designed to accurately track signals with a known structure among noise and signal components of different, “unknown,” structure. The tracking characteristics of these filters, i.e., the predicted signal amplitude versus time values versus exact signal amplitude versus time values, can be tailored to accurate tracking of harmonics buried in other signal components and noise, even at high rates of change of the reference RPM. A key to the successful construction is the precise knowledge of the structure of the signal to be tracked. For signals that vary with an axle RPM, an accurate estimate of the instantaneous RPM is essential, and procedures to this end will also be presented.

/pdf/high-resolution-order-tracking-at-extreme-slew-rates-using-3c4l3al2wo.pdf

High Resolution Order Tracking at Extreme Slew Rates Using Kalman Tracking Filters

The filter characteristics of the Vold-Kalman (1993, 1960, 1961) order tracking filter are presented. Both the frequency response as well as the time response and their time-frequency relationship have been investigated for different filter types and guidelines for optimum choice of filter parameters are presented. The Vold-Kalman filter allows for the high performance simultaneous tracking of orders in systems with multiple independent shafts. With this new filter and using multiple tacho references, waveforms, as well as amplitude and phase may be extracted without the beating interactions that are associated with conventional methods. Orders extracted as waveforms have no phase bias, and may hence be used for playback, synthesis and tailoring.

/pdf/characteristics-of-the-vold-kalman-order-tracking-filter-10ljuqf9de.pdf

Characteristics of the Vold-Kalman order tracking filter

A controller retrieves from memory a series of discrete end effector positions consisting of location and orientation of the end effector. Joint angle values are computed in real time corresponding to intermediate points along the path in a manner which provides smooth end effector motion, avoids the problems of singularities and efficiently accommodates kinematic redundancy. In particular, end effector reorientation between successive goal points is smoothly accomplished by rotating the end effector about the eigenvector of the two successive end effector orientations.

Industrial robot with controller

Theoretical Foundations for High Performance Order Tracking with the Vold-Kalman Tracking Filter

In the sine dwell school of modal testing, a frequency domain function, most often called the Mode Indicator Function (MIF), has been employed for several years to detect the presence of real normal modes. This function is calculated from Frequency Response Functions from a single exciter location to highlight those frequencies where the global response phase lags the sinusoidal excitation by 90 degrees. Recently, a multivariate mode indicator procedure was presented, that, using FRF's from several exciter locations, can detect repeated roots, as well as providing initial force patterns for the sine dwell tuning of the associated modes. This paper presents a brief outline of the theory of this multivariate mode indicator and then shows how the information generated by this procedure may be used for the extraction of selected modal parameters through function enhancement and frequency domain SDOF techniques. The use of the multivariate mode indicator in the traditional MDOF procedures will also be discussed. Examples will be presented from several tests.

Håvard Vold

Papers

High Resolution Order Tracking at Extreme Slew Rates Using Kalman Tracking Filters

Characteristics of the Vold-Kalman order tracking filter

Industrial robot with controller

Theoretical Foundations for High Performance Order Tracking with the Vold-Kalman Tracking Filter

The multivariate mode indicator function in modal analysis