Complementary limb motion estimation based on interjoint coordination using principal components analysis

For the restitution of walking of a hemiplegic patient by means of a motorized orthosis, as well as in intelligent prosthetics, a major challenge is the coordination of healthy and robotically assisted limbs. The approach suggested here employs the method of Principal Components Analysis to first analyze the coupling of human Degrees of Freedom (DoFs) in healthy subjects. Based on this knowledge, adequate motion for inoperable DoFs in impaired patients is estimated on-line from sound limb motion. Thus, the intention of a partially paralyzed person or an amputee can be deduced from residual body motion, in order to coordinately actuate or supervise the impaired limbs. To evaluate the approach, simulations with recorded gait trajectories of healthy subjects are performed. The results of these theoretical investigations show a promising potential.