Coupled oscillators utilised as gait rhythm generators of a two-legged walking machine

Abstract

The gait of current two-legged walking machines differs from that of humans, although the kinematic structures of these machines' legs frequently imitate human limbs. This paper presents a method of generating the trajectories of hip and knee joint angles resulting in a gait pattern similar to that of a human. For this purpose the solutions of coupled van der Pol oscillator equations are utilised. There is much evidence that these equations can be treated as a good model of the central pattern generator generating functional (also locomotional) rhythms in living creatures. The oscillator equations are solved by numerical integration. The method of changing the type of gait by changing appropriate parameter values in the oscillator equations is presented (change of velocity and trajectory of leg-ends). The results obtained enable enhanced control of two-legged walking systems by including gait pattern generators which will assume a similar role to that of biological generators.

Reconfigurable Double Inverted Pendulum Applied to the Modelling of Human Robot Motion

“…There is a variety of methods used, for example: using captured movement of human subject and motion generation based on trajectories produced by Central Patterns generators: [18], [6]. In those methods trajectories of every joint are generated, taking into account the structure of the whole robot.…”

Análisis teórico y computacional de la marcha normal y patológica: una revisión

“…En los últimos años ha habido un creciente interés en este tipo de modelos (35,75,77), por el interés en mejorar la comprensión de las interacciones en los sistemas neuromusculares, permitiendo simular la marcha sobre la base neuro-fisiológica de la GPC y la biomecánica del aparato locomotor humano y de la compleja interacción entre el sistema nervioso y el sistema músculo-esquelético (35). Taga y col. (71,72), han propuesto un modelo neuro-músculo-esquelético con el cual se exploran los mecanismos de control, encargados de generar el movimiento humano sobre la base de las propiedades emergentes del GCP como sistema neuronal básico.…”

CPG Implementations for Robot Locomotion: Analysis and Design

“…In spite of its importance, the CPG-approach presents several disadvantages. One of the main drawbacks of CPGs is related with the learning methodologies to generate the rhythmic signals (Zielinska, 1996). For that purpose, a method to tune a CPG using genetic algorithms (GA) is proposed, whose details are provided in section 5.…”