• Transactions on Control, Automation and Systems Engineering
• /
• v.4 no.2
• /
• pp.124-129
• /
• 2002

This paper demonstrates that the largest Lyapunov exponent λ of recurrent neural networks can be controlled efficiently by a stochastic gradient method. An essential core of the proposed method is a novel stochastic approximate formulation of the Lyapunov exponent λ as a function of the network parameters such as connection weights and thresholds of neural activation functions. By a gradient method, a direct calculation to minimize a square error (λ - λ$\^$obj/)$^2$, where λ$\^$obj/ is a desired exponent value, needs gradients collection through time which are given by a recursive calculation from past to present values. The collection is computationally expensive and causes unstable control of the exponent for networks with chaotic dynamics because of chaotic instability. The stochastic formulation derived in this paper gives us an approximation of the gradients collection in a fashion without the recursive calculation. This approximation can realize not only a faster calculation of the gradient, but also stable control for chaotic dynamics. Due to the non-recursive calculation. without respect to the time evolutions, the running times of this approximation grow only about as N$^2$ compared to as N$\^$5/T that is of the direct calculation method. It is also shown by simulation studies that the approximation is a robust formulation for the network size and that proposed method can control the chaos dynamics in recurrent neural networks efficiently.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.18 no.2
• /
• pp.380-388
• /
• 1994

Study on the chaotic stirring has been performed numerically and experimentally for a shallow rectangular tank accompanying a vortex shedding. The model is composed of a rectangular tank with a vertical plate with a length half the width of the tank. The tank is subject to a horizontal sinusoidal oscillation. The chaotic stirring was analysed by Poincare sections, unstable manifolds and Lyapunov exponents. As Reynolds number is increased the stirring effect is decreased due to the growth of a regular regions near the lower surface of the tank. In the other hand decrease of Reynolds number gives a weaker vortex shedding resulting in the poorer stirring effect. It was also found that the Lyapunov exponent is the highest at the dimensionless period of 1.3-1.5, which seems to be the best condition for the efficient stirring. The experimental visualization for the deformation of materials exhibits the striation pattern similar to the unstable manifold obtained numerically.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1844-1847
• /
• 1997

Assuming that EEG(electroencephalogram), which is generated by a nonlinear electrical of billions of neurons in the brain, has chaotic characteristics, it is confirmend by frequency spectrum analysis, log frequency spectrum analysis, correlation dimension analysis and Lyapunov exponents analysis. Some chaotic characteristics are related to the degree of brain activity. The slope of log frequency spectrum increases and the correlation dimension decreasess with respect to the activities, while the largest Lyapunov exponent has only a rough correlation.

• Journal of the Chungcheong Mathematical Society
• /
• v.25 no.3
• /
• pp.589-597
• /
• 2012

In this paper we investigate the stability of solutions of the perturbed differential equations with the positive order of the perturbation by using the notion of the Lyapunov exponent of unperturbed equations and an integral inequality of Bihari type.

• Journal of Biomedical Engineering Research
• /
• v.15 no.3
• /
• pp.237-244
• /
• 1994

We investigated the qualitive and quantitative properties in EEG signal which responds to auditory stimulus with increaing the sound-cutting frequency from 2 Hz to 20 Hz by 2 Hz step units, by chaotic dynamics. To bigin with, general chaotic properties such as fractal mechanism, 1 If frequency spectrum and positive Lyapunov exponent are discussed in EEG signal. For evoked potential with given auditory stimulus, the route to chaos by bifurcation diagram and the changes in geometrical property of Poincare sections of 2-dimensional psedophase space is observed. For that containing spontaneous potential, seen as the random background signal, the chaotic attractors in 3-dimensional phase space are found containing the same infomation as the above mentioned evoked potential. Finally the chinges of Lyapunov exponent by various sound-cutting frequencies of stimulus and by the various spatial positions (occipital region) in a brain surface to be measured, are illustrated meaningfully.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09b
• /
• pp.110-113
• /
• 2003

In this paper, we propose that the chaotic behavior analysis in the mobile robot of embedding Arnold equation with obstacle. In order to analysis of chaotic behavior in the mobile robot, we apply not only qualitative analysis such as time-series, embedding phase plane, but also quantitative analysis such as Lyapunov exponent in the mobile robot with obstacle. In the obstacle, we only assume that all obstacles in the chaos trajectory surface in which robot workspace has an unstable limit cycle with Van der Pol equation.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09b
• /
• pp.5-8
• /
• 2003

In this paper, we propose that the chaotic behavior analysis in the mobile robot of embedding Chua's equation with obstacle. In order to analysis of chaotic behavior in the mobile robot, we apply not only qualitative analysis such as time-series, embedding phase plane, but also quantitative analysis such as Lyapunov exponent in the mobile robot with obstacle. In the obstacle, we only assume that all obstacles in the chaos trajectory surface in which robot workspace has an unstable limit cycle with Van der Pol equation

• International Journal of Automotive Technology
• /
• v.8 no.4
• /
• pp.467-476
• /
• 2007

This work verifies the chaotic motion of a steer-by-wire vehicle dynamic system, and then elucidates an application of dither smoothing to control the chaos of a vehicle model. The largest Lyapunov exponent is estimated from the synchronization to identify periodic and chaotic motions. Then, a bifurcation diagram reveals complex nonlinear behaviors over a range of parameter values. Finally, a method for controlling a chaotic vehicle dynamic system is proposed. This method involves applying another external input, called a dither signal, to the system. The designed controller is demonstrated to work quite well for nonlinear systems in achieving robust stability and protecting the vehicle from slip or spin. Some simulation results are presented to establish the feasibility of the proposed method.

• Proceedings of the KSME Conference
• /
• 2007.05a
• /
• pp.792-795
• /
• 2007

To quantify irregular body motions the time series analysis was applied to the gait study. The motions obtained from gait experiment are complex to exhibit nonlinear behaviors. The purpose of this study is to measure quantitatively the characteristics of the major six joints of the body during walking. The gait experiments were carried out for eighteen young males walking on a motor driven treadmill. Joint motions were captured using eight video cameras, and then three dimensional kinematics of the neck and the upper and lower extremities were computed by KWON 3D motion analysis software. The largest Lyapunov exponent was calculated from the time series to quantify stabilities of each joint. The results provides a data set of nonlinear dynamic characteristics for six joints engaged in normal walking.

• Bulletin of the Korean Mathematical Society
• /
• v.51 no.2
• /
• pp.457-478
• /
• 2014

In this paper, a 3D autonomous system, which has only stable or non-hyperbolic equilibria but still generates chaos, is presented. This system is topologically non-equivalent to the original Lorenz system and all Lorenz-type systems. This motivates us to further study some of its dynamical behaviors, such as the local stability of equilibrium points, the Lyapunov exponent, the dissipativity, the chaotic waveform in time domain, the continuous frequency spectrum, the Poincar$\acute{e}$ map and the forming mechanism for compound structure of its special cases. Especially, with the help of the Project Method, its Hopf bifurcation of codimension one is in detailed formulated. Numerical simulation results not only examine the corresponding theoretical analytical results, but also show that this system possesses abundant and complex dynamical properties not solved theoretically, which need further attention.