How do I prove that the given system is globally asymptotically stable, using Lyapunov analysis?

\begin{equation} \left.\begin{aligned} \dot{x_1} &= x_2 \\ \dot{x_2} &= -\frac{x_1}{1 + x_2^2} \label{eq:q2} \end{aligned}\qquad\right\} \end{equation}

  • $\begingroup$ Are you familiar with Lyapunov functions? And what have you tried yourself? $\endgroup$ – Kwin van der Veen Nov 12 '18 at 16:58
  • $\begingroup$ I have tried general quadratic functions, such as (x1^2 + x2^2), but a term xy remains and I can't conclude properly. $\endgroup$ – Abhinav Sinha Nov 12 '18 at 17:25

You could try to get the Lyapunov function in the form $$ V=\frac12x_1^2+g(x_2) $$ so that $$ \dot V=x_1x_2+g'(x_2)\frac{-x_1}{1+x_2^2} $$ so that one would get a usable result with $g'(x_2)=x_2(1+x_2^2)$, integrating to,for example, $g(x_2)=\frac14(1+x_2^2)^2$.

  • $\begingroup$ Thank you so much. I was taking something similar to g(x2) but it was constant and I was stuck. Thanks for your help :) $\endgroup$ – Abhinav Sinha Nov 12 '18 at 18:55
  • 1
    $\begingroup$ @SampleTime : Then consider $V-V(0)$ to get the normalized value. I prefer the completed square for its simplicity. $\endgroup$ – LutzL Nov 12 '18 at 21:18
  • 1
    $\begingroup$ @SampleTime : Could you elaborate? How to get a negative value from $V-V(0)=\frac12x_1^2+\frac12x_2^2+\frac14x_2^4$? $\endgroup$ – LutzL Nov 12 '18 at 21:27
  • $\begingroup$ @LutzL You are correct, nice! I made an error when writing down the equation, sorry for the confusion. $\endgroup$ – SampleTime Nov 12 '18 at 21:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.