Search Results

This pilot study investigated the potential of using trunk acceleration feedback control of center of pressure (COP) against postural disturbances with a standing neuroprosthesis following paralysis. Artificial neural networks (ANNs) were trained to use three-dimensional trunk acceleration as input to predict changes in COP for able-bodied subjects undergoing perturbations during bipedal stance. Correlation coefficients between ANN predictions and actual COP ranged from 0.67 to 0.77. An ANN trained across all subject-normalized data was used to drive feedback control of ankle muscle excitation levels for a computer model representing a standing neuroprosthesis user. Feedback control reduced average upper-body loading during perturbation onset and recovery by 42% and peak loading fby 29% compared with optimal, constant excitation.

Feedback control of a human-powered single-track bicycle is investigated through the use of a linearized dynamical model in order to develop feedback gains that can be implemented by a human pilot in an actual vehicle. The object of the control scheme is to satisfy two goals: balance and tracking. The pilot should be able not only to keep the vehicle upright but also to direct the forward motion as desired. The two control inputs, steering angle and rider lean angle, are assumed to be determined by the rider as a product of feedback gains and “measured” values of the state variables: vehicle lean, lateral deviation from the desired trajectory, and their derivatives. Feedback gains are determined through linear quadratic regulator theory. This results in two control schemes, a “full” optimal feedback control and a less complicated technique that is more likely to be usable by an inexperienced pilot. Theoretical optimally controlled trajectories are compared with experimental trajectories in a lane change maneuver.

A simple pendulum model is used to study how feedforward and feedback can be combined to control rhythmic limb movements. I show that a purely feedforward central pattern generator (CPG) is highly sensitive to unexpected disturbances. Pure feedback control analogous to reflex pathways can compensate for disturbances but is sensitive to imperfect sensors. I demonstrate that for systems subject to both unexpected disturbances and sensor noise, a combination of feedforward and feedback can improve performance. This combination is achieved by using a state estimation interpretation, in which a neural oscillator acts as an internal model of limb motion that predicts the state of the limb, and by using alpha-gamma coactivation or its equivalent to generate a sensory error signal that is fed back to entrain the neural oscillator. Such a hybrid feedforward/feedback system can optimally compensate for both disturbances and sensor noise, yet it can also produce fictive locomotion when sensory output is removed, as is observed biologically. CPG behavior arises due to the interaction of the internal model and a feedback control that uses the predicted state. I propose an interpretation of the neural oscillator as a filter for processing sensory information rather than as a generator of commands.

This study compared the nature of postural adaptations and control tendencies, between 7 (n = 9) and 11-year-old boys (n = 10) with Developmental Coordination Disorder (DCD) and age-matched, younger (n = 10) and older (n = 9) peers in a leaning task. Examination of anterior-posterior, medio-lateral, maximum and mean area of sway, and path length revealed one significant interaction as older, unaffected boys swayed more than all other groups (p < .01). As a group, boys with DCD displayed smaller anterior-posterior (p < .01) and area of sway (p < .01). Analysis of relative time spent in the corrective phase (p < .002) revealed that boys with DCD spent 54% under feedback control while boys without DCD spent 78%. This was attributed to reduced proprioceptive sensitivity, as confirmed by significant differences between the groups (p < .009) in spectral analysis of peak frequency of sway.

Two experiments are reported that focus on manipulating both the context and the spatial precision of a computer-pointing task. Single goal-directed actions are compared to dual-phase tasks, where participants are required to sequentially attain two goal locations. Results support the idea that for movements in series, movement planning, and online feedback, control can occur simultaneously. Additionally, for single-phase tasks and the final phase of dual-phase tasks, the termination requirement influences the temporal components of the movement. The effects of termination and movement context appear to hold regardless of the spatial precision of the task. This suggests that the effects of spatial precision and movement termination are independent, although both have an impact on the deceleration time for goal-directed movements.