A comparative study of computation of Lyapunov spectra with different algorithms (Q1972886)

The authors are interested in the computation of Lyapunov exponents for a general system of ordinary differential equations \[ {dZ\over dt} =F(Z,t). \] They consider three numerical procedures. One is standard (it is based on a reorthogonalization algorithm); the second one is a differential version of the previous one; the third one has been recently proposed and it is based on a matrix decomposition. The performances of the three methods are compared and discussed for some low-dimensional typical examples: van der Pol oscillator, Lorenz system, anisotropic Kepler problem, coupled oscillators.