Tadao Kawai

A new method for identifying rigid link models of lower limbs

A new method for identifying rigid link models of human lower limbs has been proposed in this paper. The method was motivated by necessity of simulating human body movements for rehabilitation or for design of assistive devices. The method is based on combination of random search and least squares estimation techniques. Simulation and experimental results are given to illustrate the effectiveness of the proposed method.

Parameter Identification of Beam Using Image Processing (6th Report, Improvement of Parameter Identification Using Orthogonal Functions)

In this paper, we improved previously proposed parameter identification technique using orthogonal functions to identify the density and the modulus of elasticity of each segment of the beam. Normalizing the parameters by the designed value and singular value decomposition technique were applied to the identification process. For the poposed technique well regulated the instability in the inverse ploblem, we could apply the Haar series for any type of beam with large discountinuity. We carried out simulation and showed good applicability in this paper.

Parameter identification of Beam using image processing 2nd report, in the case of non-uniform beam

Publisher Summary This chapter evaluates the efficacy of several techniques for identifying the density and the modulus of elasticity in a small portion of a beam. Parameters of a finite element model, i.e. the density and the modulus of elasticity, of two types of cantilever beam are estimated using an image-processing technique. One beam is steel combined with brass and the other is a damaged beam with a small crack. The image-processing technique is known to have significant advantages over the traditional measuring technique. That is, image processing is a non-destructive, non-contact, and variable zooming step technique. To identify the parameters of a small segment of the combined beam, Spline smoothing with symmetrization, and the weighted Tikhonov regularization technique are used. In addition, to detect the size and the location of a small crack, a new type of simulated annealing method is proposed. Both methods studied are shown to yield highly accurate estimations.

Detection of Regenerative Chatter Using Spectral Analysis

On regenerative chatter occurring in turning workpieces, a method of detecting chatter is proposed that requires only one threshold value. The frequency characteristic of a workpiece during cutting was investigated by using spectral analysis first, and then the algorithm of detection was thought out from the investigated results. This algorithm is based on the regenerative chatter theory. Experiments varying cutting speed and width of cut on orthogonal cut were carried out, and this method was found available. Furthermore, it was shown that the present method can also detect regenerative chatter from dynamic cutting force.