Date of Award

Fall 2010

Document Type


Degree Name

Doctor of Philosophy (PhD)


Biomedical Engineering

First Advisor

Alan W. L. Chiu


The loss of a limb tremendously impacts the life of the affected individual. In the past decades, researchers have been developing artificial limbs that may return some of the missing functions and cosmetics. However, the development of dexterous mechanisms capable of mimicking the function of the human hand is a complex venture. Even though myoelectric prostheses have advanced, several issues remain to be solved before an artificial limb may be comparable to its human counterpart. Moreover, the high cost of advanced limbs prevents their widespread use among the low-income population.

This dissertation presents a strategy for the low-level of control of a cost effective robotic hand for prosthetic applications. The main purpose of this work is to reduce the high cost associated with limb replacement. The presented strategy uses an electromyographic signal classifier, which detects user intent by classifying 4 different wrist movements. This information is supplied as 4 different pre-shapes of the robotic hand to the low-level of control for safely and effectively performing the grasping tasks. Two proof-of-concept prototypes were implemented, consisting on five-finger underactuated hands driven by inexpensive DC motors and equipped with low-cost sensors. To overcome the limitations and nonlinearities of inexpensive components, a multi-stage control methodology was designed for modulating the grasping force based on slippage detection and nonlinear force control.

A multi-stage control methodology for modulating the grasping force based on slippage detection and nonlinear force control was designed. The two main stages of the control strategy are the force control stage and the detection stage. The control strategy uses the force control stage to maintain a constant level of force over the object. The results of the experiments performed over this stage showed a rising time of less than 1 second, force overshoot of less than 1 N and steady state error of less than 0.15 N. The detection stage is used to monitor any sliding of the object from the hand. The experiments performed over this stage demonstrated a delay in the slip detection process of less than 200 milliseconds. The initial force, and the amount of force incremented after sliding is detected, were adjusted to reduce object displacement. Experiments were then performed to test the control strategy on situations often encountered in the ADL. The results showed that the control strategy was able to detect the dynamic changes in mass of the object and to successfully adjust the grasping force to prevent the object from dropping.

The evaluation of the proposed control strategy suggests that this methodology can overcome the limitation of inexpensive sensors and actuators. Therefore, this control strategy may reduce the cost of current myoelectric prosthesis. We believe that the work presented here is a major step towards the development of a cost effective myoelectric prosthetic hand.