Steven W. Peterson

An anthropomorphic hand exoskeleton to prevent astronaut hand fatigue during extravehicular activities

This correspondence presents a prototype of a powered hand exoskeleton that is designed to fit over the gloved hand of an astronaut and offset the stiffness of the pressurized space suit. This will keep the productive time spent in extravehicular activity from being constrained by hand fatigue. The exoskeleton has a three-finger design, the third and fourth fingers being combined to lighten and simplify the assembly. The motions of the hand are monitored by an array of pressure sensors mounted between the exoskeleton and the hand. Controller commands are determined by a state-of-the-art programmable microcontroller using pressure sensor input. These commands are applied to a PWM driven dc motor array which provides the motive power to move the exoskeleton fingers. The resultant motion of the exoskeleton allows the astronaut to perform both precision grasping tasks with the thumb and forefinger, as well as a power grasp with the entire hand.

Simultaneous measurement of body center of pressure and center of gravity during upright stance. Part I: Methods

Abstract Clinical and experimental assessments of balance use measures of center of pressure (COP) excursion to quantify postural stability during standing. This assumes that the greater the COP excursions, the greater the imbalance. However, it is the position of the body center of gravity (COG), specifically in relation to the base of support, that determines static stability during upright standing. The reason the COG is frequently ignored is that it cannot be measured directly, whereas the COP is readily obtained from a force platform. We report here on a method for simultaneous measurement of the COP and COG displacements. To compute the COG displacements, an optoelectric imaging system operating in synchrony with a force platform was used to measure the three-dimensional positions of the body joints and body segment endpoints. The COG displacements were then computed from the segment kinematics using subject-specific anthropometric measurements. In Part II of this paper, we compare summary measures of the COP and COG excursions obtained from six subjects during three different upright stances.

Simultaneous measurement of body center of pressure and center of gravity during upright stance. Part II: Amplitude and frequency data

Abstract In this paper (Part II) we present amplitude and frequency data on simultaneously measured excursions of the body center of gravity (COG) and center of pressure (COP). The excursions were measured from six subjects during 10-s trials of three upright postural stances: standing on both legs or double stance with eyes open (DSEO), double stance with eyes closed (DSEC), and standing on one leg or single stance with eyes open (SSEO). This study represents the first attempt to describe the normative characteristics and stance-dependent differences between the anteroposterior and mediolateral components of both the COP and COG signals, measured from multiple individuals by the technique described in Part 1. We quantified the anteroposterior and mediolateral excursions of the COP and COG using amplitude and frequency measures, and their planar excursions using 95% confidence ellipses. Although the COP-based amplitude measures were significantly greater than their COG-based counterparts, the amplitude and frequency measures of both were highly correlated in all stances. The results support the continued use of COP-based measures to quantify impaired balance during upright stance.

DYNAMIC ANALYSIS OF SPACE-BASED INFLATED BEAM STRUCTURES

This paper examines the dynamic behaviors of inflated beam aerospace structures. The principal foci of this investigation are the determination of the damping mechanisms active in structures constructed from inflated cylindrical beams, development of a practical modeling method for complex structures, and an examination of the difficulties in predicting on‐orbit dynamic behavior from ground tests. A Euler–Bernoulli model of the inflated beam is used to determine distributed damping coefficients from modal tests. The results show that the viscous damping in the inflated beam is independent of beam pressure, but that the pressurization stress levels in the beam fabric affect the strain‐rate damping. The Euler–Bernoulli inflated beam model is used in conjunction with a conventional finite element package to model the dynamic behavior of a complex inflated beam structure, a 1.7‐m‐diam inflated dish antenna mockup. The model accurately predicts the lower natural frequencies of the dish structure. A comparison ...

Reaction force patterns of injured and uninjured knees during walking and pivoting

A precise knowledge of the biomechanical alterations produced by anterior cruciate ligament (ACL) damage would aid in selecting appropriate therapeutic intervention and monitoring rehabilitation. In an attempt to assess dynamic knee joint function, we compared the ground reaction force (GRF) and vertical couple patterns during walking and pivoting from an ACL-deficient population with those from a separate uninjured population. Statistical methods were used to quantitate the differences between the two populations for each force pattern in each functional task and to delineate the intervals of the force patterns in each functional task during which significant differences existed between the two populations. Our results indicate that significant differences exist between the GRF and vertical couple patterns of ACL-deficient subjects and uninjured subjects, but that onset and duration of these differences during stance phase vary among force components and tasks. The processing scheme extracts significant differences in the GRF and vertical couple patterns that would be lost in a comparison of a few pattern descriptors. Our results suggest that the vertical couple should be evaluated during assessment of pivoting maneuvers and show promise of providing useful information for assessment of knee dysfunction.

Power Assist EVA Glove Development

Structural modeling of the EVA glove indicates that flexibility in the metacarpophalangeal (MCP) joint can be improved by selectively lowering the elasticity of the glove fabric. Two strategies are used to accomplish this. One method uses coil springs on the back of the glove to carry the tension in the glove skin due to pressurization. These springs carry the loads normally borne by the glove fabric, but are more easily deformed. An active system was also designed for the same purpose and uses gas filled bladders attached to the back of the EVA glove that change the dimensions of the back of the glove and allow the glove to bend at the MCP joint, thus providing greater flexibility at this joint. A threshold control scheme was devised to control the action of the joint actuators. Input to the controller was provided by thin resistive pressure sensors placed between the hand and the pressurized glove. The pressure sensors consist of a layer of polyester film that has a thin layer of ink screened on the surface. The resistivity of the ink is pressure dependent, so an extremely thin pressure sensor can be fabricated by covering the ink patch with another layer of polyester film and measuring the changing resistance of the ink with a bridge circuit. In order to sense the force between the hand and the glove at the MCP joint, a sensor was placed on the palmar face of the middle finger. The resultant signal was used by the controller to decide whether to fill or exhaust the bladder actuators on the back of the glove. The information from the sensor can also be used to evaluate the effectiveness of a given control scheme or glove design since the magnitude of the measured pressures gives some idea of the torque required to bend a glove finger at the MCP joint. Tests of this actuator, sensor, and control system were conducted in an 57.2 kPa glove box by performing a series of 90 degree finger bends with a glove without an MCP joint assembly, a glove with the coil spring assembly, and with the four fingered actuated glove. The tests of these three glove designs confirm the validity of the model.

Design and structural analysis of highly mobile space suits and gloves.

This paper evaluates the factors that control the flexibility of fabric space-suit elements, in particular gloves, by examining a bending model of a pressurized fabric tube. Results from the model are used to evaluate the design strategies used in space-suit components, to evaluate the current direction in research on highly mobile space-suit gloves and to suggest changes necessary for optimum glove fabric selection. Finally it is shown that the modulus of the fabric used in space-suit joint construction is as important to the flexibility of the joint as the glove size and design.

A Prototype Power Assist EVA Glove

The most recent generation of space suit EVA gloves has addressed the problem of loose fit and stiffness in the fingers, but it remains difficult to build a glove assembly with low metacarpophalangeal joint stiffness. Fatigue due to constantly displacing the glove from a neutral position has been reported as the limiting factor in some EVA activities. This paper outlines an actuation system that uses gas filled bladders attached to the back of the EVA glove to provide the necessary force to bend the glove at the metacarpal joint, thus providing greater endurance during finger grasping tasks. A simple on-off controller senses hand movement through small pressure sensors between the finger and the glove restraint. The controller then fills or exhausts the bladders on the back of the glove to effectively move the neutral position of the glove as the hand inside moves.

Highly mobile space suit material optimization

This paper discusses the factors that control the flexibility of fabric space suit elements by examining a bending model of a pressurized fabric tube. Results from the model are used to evaluate the current direction in highly mobile EVA glove research and suggest that changes are necessary in the suit and glove fabric selection methodology.

Simulation of a thermoelectric element using B-spline collocation methods

Many authors have described ways of calculating the efficiency of thermoelectric devices. These calculations usually make simplifying assumptions to define the behavior of the device. The most common of these is to make the thermoelectric material properties constant. This paper presents work in progress toward simulating the behavior of real thermoelectric devices. We present the steady state differential equations defining the behavior of thermoelectric materials and the associated boundary conditions required to solve power generation and cooling element problems. We implemented a simple one-dimensional simulation of a power-generating device using B-spline collocation methods. The results of the simulation are compared to an example presented by Angrist (1976).