Improving the parameters of neural oscillators to generate joint trajectories of an exoskeleton for lower limbs

This work deals with an optimization system developed to find optimal parameters for neural oscillators used to generate joint trajectories of an exoskeleton for lower limbs. The exoskeleton is considered as a biped robot that presents cyclical joint trajectories during the walking. Matsuoka neural oscillator, which consists of two mutual inhibitory neurons and is modeled by two differential equations for each one, is being used as trajectory generator for the robot joints. The neural oscillators are able to produce a cyclical output. However, the parameters of the differential equations are difficult to be set for a given desired output. Thus, we have implemented an optimization system to find the better values to the neural oscillator parameters given a predefined desired joint trajectory of an exoskeleton for lower limbs. This optimization system works to minimize the error between the trajectory generated by the oscillator and the desired trajectory, regarding the robot dynamics. The advantage of using oscillators is justified because the trajectories can be generated in real time, with a less time-consuming with relation to the analytical methods. A simulation of the exoskeleton considering the optimization system is presented. The results show the proposed optimization system and the trajectory generator using neural oscillators can be applied in an adaptive model that include interaction forces between the user and the robot, so changing the trajectory according to the user intention.