Study on Optimization of CPG Model for Lower Limb Prosthesis

Abstract:

A Multitude of different prosthesis designs have been developed for restoring transtibial and transfemoral amputees’ mobility.But yet,most of them are considered as passive devices.Therefore,more and more researchers develop bionic controller that simulate the biological certral pattern generator,namely CPG.This paper presents a new control method using bipedal robotics technology and bio-inspiration based on CPG.To begin with,we present the fundamental measurement of human walking gait and the device mainly includes three-dimensional camera system,digitized movements analyzer and so on.We choose hopf oscillators as CPG simulation unit.And after several tests,five oscillators are just right for a single joint.We change the simple hopf oscillator equation.Thus corresponding to the knee,CPG modeling finally generate actual human angle curve.Then we define learning equation as learning a given periodic signal.By trying different values of the different parameters we obtain the desired walking curve of knee joint.Using the obtained parameters,learning equation reproduce knee joint angle.According to the signal of the accelerometer that placed in the hip to adjust learning equation,so the amputees can easily control the speed of walking.Matlab simulation results show that the same trend with changes in human joint angles,which lay a good foundation for the control of active prostheses.

• Periodical: Applied Mechanics and Materials (Volume 461)
• Pages: 633-640
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.461.633
• Online since: November 2013
• Authors: Xin Guo, Ming Yue Li, Cai Yu Xu
• Keywords: Central Pattern Generator (CPG), Hopf Oscillator, MATLAB Simulation, Walking Gait

References

[1] Verdaasdonk B W, Koopman H F J M, Van Der Helm F C T, Energy efficient and robust rhythmic limb movement by central pattern generators, Neural Networks, vol. 19, no. 4, pp.388-400, (2006).

DOI: 10.1016/j.neunet.2005.09.003

[2] Grillner S, Neurobiological bases of rhythmic motoracts in vertebrates, Science, vol. 228, pp.143-149, (1985).

DOI: 10.1126/science.3975635

[3] G.N. Orlovsky, TG Deliagina, and S. Grillner, Neuronal control of locomotion: from mollusc to man, Oxford University Press USA, 1999. 2.

DOI: 10.1016/s1388-2457(00)00328-x

[4] JJ Collins and I. Stewart, Hexapodal gaits and coupled nonlinear oscillator models, Biological Cybernetics, vol. 68 no. 4, pp.287-298, 1993. 2.

DOI: 10.1007/bf00201854

[5] Allahabad, Bio-inspired control methodology of walking for intelligent prosthetic knee[Master thesis], Computer Sciences, Indian Institute Of Information Technology, (2005).

[6] G.C. Nandi, A.J. Ijspeert, P. Chakraborty, Anirban Nandi. Development of adaptive modular active leg (AMAL) using bipedal robotics technology, Robotics and Autonomous Systems, (2009).

DOI: 10.1016/j.robot.2009.02.002