Dynamics, stability, and control of stepping.

The dynamics, stability, and control of stepping are considered. The role of internal models is elaborated. The main objective of the paper is to provide a better understanding of the machinery and processing in the central nervous system (CNS) that relates to stepping. The role of the vestibular system in balance and balance recovery is described. Balance and balance recovery are essential in stepping, and guarantee the stability of the system before, during, and after stepping. In sagittal standing, humans use two distinct sets of control strategies to maintain their postural stability in response to external disturbance. In one set of strategies, the configuration of the base of support, namely, the position of the feet, remains unchanged. The ankle and hip strategies are examples of postural adjustments where the feet do not move. When the disturbances are large, and move the center of mass or pressure outside the support boundaries, stepping strategies are required. A simple control strategy is proposed for illustrative purposes. Its effectiveness is verified by computer simulation of a seven-link two-dimensional sagittal biped. The applications of the model in assessing trauma and injury are discussed.

Rhythmic Movement of a Pair of One-Link Arms: Coordination by Intermittent Control.

This paper considers the coordination and control of periodic movements of a pair of one-link arms. The system consists of two one-link arms each controlled by two muscle-like actuators. The muscle-like actuators are activated by simulated neural inputs. The model is simple enough to analyze, yet it embodies many aspects of human arms. Three attributes of the rhythmic coordinated movement of two arms, namely frequency, magnitude, and relative phase, are the only inputs to the controller. The controller uses mild co-activation and primarily activates the agonist. The effects of transmission delays, present in the reflex loop of physiological systems, also are modeled. The results of this research indicate the feasibility of controlling oscillatory body movements with short periods of activation. The result of many simulations, by varying the frequency or amplitude of the movement, indicate that the apparent lack of a simple relationship between neural control and desired behavior of the system should not be mistaken as evidence for the absence of a causal relationship between the activation patterns of the muscles and the desired behavior. Simulations of this system show stable oscillations at different frequencies and magnitudes even with additive noise and changes in the system mass.