Approximating the stance of a 2-DOF monoped runner (Q1840684)

The authors investigate legged locomotion of animals and robots. They specifically study the stance dynamics through a general Hamiltonian model of a spring law without taking the effect of gravity into consideration. Although the model is integrable, the solution cannot in general be written with the help of known functions so they use a mean value theorem and represent the solution in a closed form with elementary functions. They compare this model with other simplified spring models and show that the error is relatively small. They also consider the perturbed locomotion including gravity. The solution found through the mean value theorem is used as the initial solution of an iterative approximation of the solutions of the perturbed model, since the last is non-integrable. Comparison of their results to simulation results of the perturbed model shows that, as the iterations increase, the mean error decreases rapidly.