A stiff tendon neuromusculoskeletal model of the knee

Now more than ever, progresses in information technology applied to rehabilitation robotics give new hopes to people recovering from different kinds of diseases and injuries. Beside the standard application of EMG signals to analyze disabilities or to track progress in rehabilitation, more focus has been put on controlling robot arms and exoskeletons. In recent years, biomechanists have developed very complex neuromusculoskeletal (NM) models of human joints to understand how the nervous system controls muscles and generates movements. Aware of these potentials, we have started a process of simplification to obtain a NM model suitable for the real-time control for a lower extremity exoskeleton. In this paper we present the NM model for the knee previously developed by Lloyd et al. We then investigate the effects of assuming the tendon infinitely stiff and show how this simplification does not affect the capacity of the model to predict muscle force and joint moment. We also assess the decrease in processing time required to calibrate the model and perform runtime estimates of muscles. Finally, we illustrate the implications of our research for the health care economic and social systems.