i. wheelchair technology tasks - university of …€¦ · 4 rerc on wheelchair technology i....
Post on 07-Sep-2018
218 Views
Preview:
TRANSCRIPT
4 RERC ON WHEELCHAIR TECHNOLOGY
I. WHEELCHAIR TECHNOLOGY TASKS
♦PM-1 Improved Electric and Electromechanical Systems
♦PM-1 a Computer Simulation Electromechanical Systems
♦PM-1 b Power Wheelchair Batteries
♦PM-1 c Power Wheelchair Controllers
♦PM-1 d Improved Wheelchair Motor Drives
♦PM-2 Advanced Materials and Mechanisms
♦PM-3 Improved User Input Devices and Control Concepts
♦PM-4 Integration of Improved Mobility Components (removed from work program)
♦PM-5 The Use of Integrated Controls by Persons with Physical Disabilities
♦PM-6 New Concepts in Powered Mobility
♦PM-7 Powered Mobility Simulator
♦MM-1 Structural Improvements to Manual Wheelchairs
♦WP-1 Consumer Responsive Mobility Prescription Process
♦WP-2 Wheelchair Prescription Software Project
♦STD-1 Research in Support of Wheelchair Standards
5FINAL REPORT: 1993-1998
Approach and Background
The research approach taken was to address each
of the major electromechanical components of the
powered wheelchair independently within Tasks PM-
1-3. Task PM-4 was intended to integrate the
component results into a complete ‘idealized’ system.
Task PM-1 investigates each of the major drive system
components (batteries, controller, motors, and power
train) as interrelated sub-tasks.
The four sub-tasks of PM-1 are as follows:
PM-1a Electromechanical System Simulation
PM-1b New Technology for Wheelchair Batteries
PM-1c Improved Power Controllers
PM-1d Improved Wheelchair Motor Drives
The initial two years of the Task PM-1 involved a
major subcontract with Westinghouse Corporation.
Unfortunately, Westinghouse has undergone
significant downsizing and restructuring which has
lead to a reduction in resources (staff and laboratories)
that were initially available to this task. Progress was
impeded by this unforeseeable event as new resources
had to be identified and new team members brought
up to speed during Year III. By the end of Year III,
most collaborative work with Westinghouse had been
terminated and other resources identified.
PM-1ACOMPUTER SIMULATION OF
ELECTROMECHANICAL SYSTEMS
Investigators: Douglas Hobson, Dave Brienza, Fazal
Mahmood, Jonathan Evans
Rationale
Designers of powered wheelchairs have few tools
to assist in the design and development of new
powered wheelchairs. This task focuses on the
development of a computer simulation tool that can
aid designers during the decision-making process
regarding the selection of various electromechanical
components. The primary strategy is to optimize the
design towards the lowest energy consumption.
Other variables, such as tipping stability, can also be
optimized.
The initial thrust of this task was to model the
components of the wheelchair using a proprietary
simulation tool (HEAVY) developed by
Westinghouse, Inc. When it became evident that the
Westinghouse tool was not appropriate for
wheelchair simulation and considerable new code
would be required, the task was scaled back to focus
on a more limited design tool for industry. This
direction was taken at the advice of our Advisory
Board at its May 1995 meeting.
Goals
To develop a computer-based design tool to
facilitate the design of powered wheelchairs for use
by wheelchair designers.
Methods Summary
A computer program called HEAVY, originally
developed by Westinghouse Inc. when it was
involved in battery powered car research, was used
as the conceptual model for the simulation program.
Many algorithms had to be modified and others
added to make the model applicable to wheelchairs.
The original program was Fortran based therefore not
readily useable by desktop computers. Prior work
done at the University of Virginia-RERC on rolling
resistance, wind drag and power consumption was
used to make the model more applicable to
wheelchairs. Comments were solicited from industry
designers regarding the desirable outcomes of the
model. Once the simulation model was completed,
validation by actual wheelchair testing using a
prescribed test course was done to check the accuracy
of the simulation model. Refinements to the
TASK: PM-1 IMPROVED ELECTRIC AND
ELECTROMECHANICAL SYSTEMS
6 RERC ON WHEELCHAIR TECHNOLOGY
simulation tool was done over time by continued
validation testing using different types of
wheelchairs.
Outcomes Summary
Conversion to C ++ code for the simulation
program was completed. The simulation program
now includes program code to carry out the
simulations as illustrated in the following flow
diagram. First stage validation of the simulation was
completed. This was done using an on-board data
measurement/collection system while driving two
production wheelchairs through a prescribed test
course. Energy consumption was measured and
compared to the simulation results.
In order to verify the accuracy of the energy
consumption model, two powered chairs were
monitored while they completed the test track
outlined in the ISO standards (ISO 7176/6) for
determining the energy consumption and range of
powered wheelchairs. Our test course covered 200
feet with the dimensions of the rectangular track
measuring 50 feet on each side. Throughout the test,
the voltage and current were recorded using a lap
top computer, which acquired readings 200 times per
second. The results of the validation were then used
to compare the results obtained when running an
identical course in the computer simulation model.
Since we were unable to obtain the specific motor
and battery characteristics for the wheelchairs that
were tested, the program used data obtained for
motors and batteries with similar characteristics.
However, due to the power capacity of the Invacare
chair tested, the characteristics were thought to be
sufficiently different to effect the results of the
simulation program. Therefore, only the results from
the Quickie P-190 are were used.
For the Quickie P-190, the measured average
speed over the ISO test track was 5.67 ft/ second. The
distance traveled was 2000 ft. and the energy
consumed was 22.7 watt-hours. By using the ISO
guidelines for determining the range of the chair, the
range for the P-190 was determined to be 6.67 miles.
For the simulation, the motor data that was used
was the Fracmo M453-W30 24-volt DC motor. The
battery data was based on the MK 22NF Gel Battery.
Using the speed of 5.67 ft/sec. as input data, the
program then calculated the drag, the gear-box losses
and the air drag to determine the torque required to
overcome these losses at the specified motor RPM.
The drag losses calculated for P-190 was 35.5 pounds.
The wheel diameter is 12.5 inches; therefore, the
torque required to overcome these losses is 221.9 in./
lb. Use the look-up tables for the motor
characteristics, the available energy of the battery is
monitored at each 1-second interval in order to
determine the range. From the simulation, the energy
Figure 1 - Flow diagram of simulation process
• • • • • • • • ••
• • • • •
•
• • •
7FINAL REPORT: 1993-1998
used while driving through the virtual ISO test course
was 52.86 watt-hours, yielding a total range of 3.86
miles.
The results of the simulation program vs. the
validation test show that the computer model
computes a range 42% less than the actual measured
results. This difference makes the use of the
simulation program impractical at its present stage.
More work is required to determine the source of
error.
It seems that the method used for determining
the drag may be incorrect. From studying how the
air, motor, and rolling drag are calculated, it appears
that the rolling resistance equations yield a larger
value than expected (32 lbs.). Accurate battery and
motor characteristics are also necessary for precise
comparative validation. Also, the effects of caster
drag, even on a firm rectangular test course, are not
adequately addressed by the simulation model. If
these deficiencies can be corrected, it appears that the
computer model can be a useful tool in studying the
effects that different batteries, motors, mass and frame
and wheel configurations have on the range and
energy efficiency of powered wheelchairs.
Recommended Future Research and Development
The C++ code for the simulation program
functions as intended. Additional experimental work
must now be done to refine the algorithms in order
to reduce the disparity between actual and simulated
values. Initial C++ code work was also done on the
modeling for static stability. However, now that the
newly revised versions of the ISO standards for static
(Part 1) and dynamic stability (Part 2) tests have been
completed, these simulations could be added to the
battery of tests. All information on the above
algorithms, program code and energy consumption
testing will be maintained on file, at least until
January 2003. This information can be made available
to any persons seriously contemplating additional
development of this simulation tool.
Publications
Alva, P and Hobson, DA, Computer simulation of poweredwheelchair electro-mechanical systems, Proceedings of theRESNA ‘96 Annual Conference, Salt Lake City, UT, June 1996
Hobson DA (in preparation)
PM-1B POWER WHEELCHAIR BATTERIES
Investigators: David Brienza, Douglas Hobson,
Mostafa Khondukar
Collaborators:Rick Blanyer, Steve Addington
(Electrosource, Inc.)
Rationale
The key component in any electrically powered
vehicle is the battery—the heaviest, most expensive,
and least reliable system component. The need for
improved battery technology is clear. Current
technologies used and the configurations made
available are far less than optimal for wheelchair
applications. For example, the basic configuration of
lead-acid batteries [Bode, 1977] limits frame design,
space for respirators, etc. Virtually every commercial
electric vehicle, including wheelchairs, uses a lead-
acid battery. For many years, lead-acid has been the
most reliable, cost-effective, and practical battery
available. It exists in its present form due to the
billions of dollars worth of research and development
aimed at improving both the battery and the mass
production process. These efforts, which were fueled
and funded almost entirely by the automobile
industry, have led to the optimization of a lead-acid
battery with respect to economics and the task of
starting a car engine. For application in wheelchairs
[Kauzlarich, 1990; Petersen, 1986; Lavanchy, 1992], the
lead-acid battery is much less than ideal. It is heavier,
more costly, and less reliable than desired, which is
not a surprising situation considering the fact that
the lead-acid battery was not originally engineered
and developed for motive power applications.
Project Goals
Our objectives for this task were:
• Review current and developing batterytechnology and evaluate its efficacy for use inpowered wheelchair systems,
• Identify one or more candidate batterytechnologies, acquire prototypes and evaluateperformance relative to wheelchair applications,and
• Disseminate findings and facilitate technologytransfer to wheelchair manufacturers.
8 RERC ON WHEELCHAIR TECHNOLOGY
Methods and Outcomes Summary
A comprehensive review of emerging battery
technology was completed and published as an RERC
Technical Report No. 2 (Bayles, 1995) and presented
at a national RESNA Conference (Bayles et al., 1994).
As a result of that effort, one candidate battery
technology was selected to evaluate for possible
application in powered wheelchairs. That
technology—the Horizon® battery—is an advanced
lead-acid technology developed by Electrosource, Inc.
of Austin, Texas. The Horizon battery is shown along
side a standard 22NF lead-acid battery in Fig. 2.
Although other technologies were considered, the
Horizon® was selected as the best battery available
for evaluation. The potential advantages are
improved energy density, improved specific energy
and a low profile design. A test plan including bench
testing and dynamometer testing was developed. The
load cycle used for testing is a variable discharge cycle
and is intended to be representative of typical indoor
and outdoor wheelchair driving. Bench testing has
been completed. Compared to commercially available
22NF gel electrolyte, lead-acid batteries, the Horizon®
battery demonstrated a 74% increase in specific
energy (40.6 Wh/kg vs. 23.3 Wh/kg).
A meeting was organized, including technical and
marketing staff from Electrosource, representatives
from three major wheelchair manufacturers, a
representative from one scooter manufacturer, and
the RERC staff was organized. At the meeting an
introduction to the new technology and preliminary
test results were shared. The research staff has no
knowledge of any further communication between
Electrosource and the wheelchair manufacturers.
Recommended Future Research and Development
Development of new battery technology has been
progressing more slowly than was anticipated in
1993. However, we expect that significant
improvements will be achieved. For this reason,
wheelchair industry representatives are advised to
stay informed and in the development loop so that
the specific requirements of the power wheelchair
may be accommodated in the packaging of any new
and significant battery technology.
Figure 2 - Horizon (right) and standard 22NF (left) lead-acid batteries
Publications
Bayles, G. New Power Source Technologies for Electric
Wheelchairs, Technical Report #2, RERC, University of
Pittsburgh, Pittsburgh, PA 1995.
Bayles, G., Ulerich, P., Palmer, K., and Brienza, D.M., New
Battery Technology for Powered Wheelchairs, Proceedings
of the 17th Annual RESNA Conference, Nashville, TN, June
1994.
References
Bode, H., Lead-Acid Batteries, John Wiley & Sons, NY, 1977.
Kauzlarich, JJ. Wheelchair batteries II: Capacity, sizing, and
life, J Rehab Res and Devel, 1990; 27(2):163-70.
Lavanchy, C. Comparative evaluation of major brands of
lead-acid batteries, Proceedings of the 1992 RESNA
International Conference, 1992;pp.541-43.
Peterson. HA. Development of test procedures for batteries
in electric wheelchairs, Report No. 86022, Energy Research
Laboratory, Niels Bohrs Alle 25, 5230 Odense M, Denmark.
PM-1C POWER WHEELCHAIR CONTROLLERS
Investigators: David Brienza and Wonchul Nho
Collaborators:Theodore Heinrich (Westinghouse
Inc.)
Rationale and Goals
Very little innovation has occurred in the
methodology used to control the power from the
batteries to the motors, which is the job of the power
controller. The objective of this development task was
to adapt an alternating current (AC) motor controller
9FINAL REPORT: 1993-1998
technology developed for an electric automobile for
use in a wheeled mobility device.
Methods and Outcomes Summary
An AC power controller using the vector control
technique was designed. An existing design
produced by the Westinghouse Corporation for
electric vehicles (EV) was modified and updated to
fit specifications developed for powered wheelchairs.
The vector controller consists of two portions,
software and hardware. Our initial work on this task
concentrated on the hardware dedicated to the high
current output stage of the controller, the motor drive.
The role of the motor drive is to convert stored energy
in the batteries to electrical power for the motors
according to the magnitude of a control signal
generated by the controller section of the device. A
block diagram of a typical electric wheelchair power
train is shown in Figure 3 below. The design for the
power controller has been completed. The new design
of the motor drive has been enhanced as compared
to the original EV design. The power switching
integrated circuits were upgraded using IGBT devices
and important performance gains were achieved with
the addition of a dead-time generator.
The design of the dead-time generator in the
motor drive has involved the theoretical
determination of three important parameters: carrier
ratio, modulation index, and time-delay. Depending
on the values and combinations of values of these
parameters, harmonic and wave form distortions can
be significant or negligible. The effect of the
significant distortions is a reduction in efficiency and
a momentary loss of control. Distortions in the
voltage-wave form have been investigated through
simulation. Distortions were determined as a function
of carrier ratio, modulation index, and time-delay.
Optimal values that minimize the distortion for both
fundamental and harmonic components of the
voltage-wave form in the output of the motor drive
were selected for three representative operating
conditions. The results of the simulation indicate that
the modulation index must be near unity, carrier
frequency is good at 15 kHz and a time delay of 10
msec is adequate. The application of these optimal
values should allow for significant improvement in
the output wave form of the motor drive.
Original plans for this task included the
fabrication and testing of a prototype controller; these
plans were not executed.
CONTROLLER
DRIVER
INPUTDEVICE
(JOYSTICK)
BATTERY
MOTORDRIVE
MOTORDRIVE
COMPUTER
IM
IM
Figure 3 - Schematic of the prototype controller
Publications
Nho WC, Brienza DM and Boston R. The development of
and AC motor drive in power wheelchair Proceedings of
15th Annual RESNA Conference, Salt Lake City, Utah, June
7-12, 1996.
PM-1D IMPROVED WHEELCHAIR MOTOR
DRIVES
Investigators: Douglas Hobson, David Brienza
Collaborator: Jules Legal
Rationale
Advancement of powered wheelchair options is
restricted by the availability of motor drive
configurations. This task explored motor
developments and specifically, motor/drive
combinations that will open new opportunities for
alternate wheelchair designs.
This task initially focused on the potential use of
AC motors and the improvement of DC motors.
However, it quickly became evident that the size of
the wheelchair market limits the development of new
motor technology specifically for use in the
wheelchair industry. Therefore, the focus was
redirected to identify existing technologies that can
be “re-packaged” in such a manner to offer new drive
options, such as a steerable in-hub motors and gear
train combinations.
10 RERC ON WHEELCHAIR TECHNOLOGY
Project Goals
1. To improve the availability of alternate wheelchair
motors/drive systems through forming working
partnerships with Federal labs and/or motor/
gear drive developers and manufacturers,
2. To work with wheelchair manufacturers in
evaluating the feasibility of introducing new
motor/train concepts and devices into new
wheelchair designs.
Figure 4 - Schematic of powered steering for front wheeldrive wheelchair
Outcomes Summary
Information and supplier literature was collected
on available motors and gear drives, such as the
Fracmo line. Direct communication was established
with Fracmo, which was followed by a joint meeting
with the Pitt-Westinghouse team in November 1994.
As a result, several prototype motor drives were
obtained and used in tasks PM-2 and PM-6. A
conceptual design was prepared and sent to a list of
manufacturers with the goal to identifying a firm that
wished to pursue a joint development project. The
same specifications were distributed throughout the
NASA technology transfer network in an effort to
identify new sources of motor/drive technology.
Finally, the following conceptual drawings were
prepared, complete with more detailed views and
specifications on the operational characteristics
required. These drawings and their contained
specifications will be used for future communications
with prospective motor/drive manufacturers.
Rearch and Development
As will be discussed in Project PM-6 below, the
commitment of a motor development and
manufacturing company will be necessary before any
new significant motor drive options will be made
available to the wheelchair industry. As part of the
PM-6 continuation plans, SBIR funding will be sought
to allow active participation by a motor company and
a wheelchair manufacturer in this effort to provide
alternate drive systems for indoor power wheelchairs.
11FINAL REPORT: 1993-1998
Rationale
Powered wheelchair maneuverability is critically
important to many people that need to maneuver their
wheelchair in confined spaces. Most products today
use the same control strategy that was used in the
first powered wheelchair introduced by Everest and
Jennings in the mid 1950s. It relies on the independent
control of the two powered wheels, usually in the rear,
and the free motion of pivoting front caster wheels.
This task and PM6 are investigating alternate methods
for enhancing wheelchair maneuverability by
changing the fundamental manner is which the
steering is accomplished. Application of successful
findings to future products will increase the number
of environments accessible to persons using these
products.
The ability of a powered wheelchair user to
maneuver in tight spaces is closely related to the
chair’s drive and steering configuration. The most
common drive configuration, differential rear wheel
drive, consists of fixed and driven rear wheels with
front caster wheels. Direction changes are made by
individually varying the speeds of the rear wheels.
In this configuration the point about which the
wheelchair pivots lies on the line perpendicular and
running through the center of the rear wheels. The
minimum turning radius is achieved when the pivot
point is directly between the rear wheels. The
minimum space required to turn the wheelchair is
then determined by the maximum distance from that
point to any other point on the wheelchair. This is
usually the front corner of the footrests or the user’s
feet (Figure 5).
To minimize the turning radius for the rear wheel
differential drive configuration, the point between the
rear wheels must be located as close to the geometric
center of the chair as possible. Several commercially
available power chairs have achieved reduced turning
radius using this approach. Another benefit of this
approach is that a larger portion of the total weight
of the wheelchair is born by the drive wheels and
less by the caster wheels. The more weight there is
on the caster wheels, the more difficult it becomes to
change directions when caster wheels must reverse
directions and rotate through 180°. The approach,
however, causes the designer to take extraordinary
steps to provide stability. Typically, stability is
achieved by counter balancing the user’s mass over
and in front of the main drive wheels with the center
of mass of the batteries located approximately at or
just rear of the axis of the main drive wheels. It is
often necessary to provide anti-tip wheels in the rear
of the chair to avoid tipping backwards while
accelerating forward. The addition of these extra
wheels may compromise the chairs ability to climb
over low obstacles if the wheels are small or close to
the ground.
Figure 5 - Rear wheel differential drive configuration
Methods Summary
An alternate approach to minimizing the turning
radius is to steer all four wheels. Steering all four
wheels avoids the problems associated with caster
wheels yet retains minimum turning radius,
maximizes stability, provides tracking of the front and
rear wheels along the same path, and provides for
enhanced obstacle climbing capability.
TASK: PM-2 ADVANCED MECHANISMS
Investigators: Clifford Brubaker, David Brienza, Douglas Hobson
Collaborators: Jules Legal, Edmund LoPresti
front
caster wheelsin front
pivot point forminimumturning radius
fixed drivewheelsin rear
12 RERC ON WHEELCHAIR TECHNOLOGY
The challenge in designing a mechanical four-
wheel steering mechanism is to design a device with
the ability to turn each wheel through 180° while
minimizing misalignment of the wheels. Steering
linkages such as those used in automobiles owe their
simple design to the relatively small turning angles
required by that type of vehicle. For highly
maneuverable small vehicles such as wheelchairs, the
range of steering angle is much greater. Furthermore,
the wheels must maintain proper alignment over the
entire range of steering angles to avoid undesirable
wheel scrubbing when the wheelchair turns. The
wheels are properly aligned whenever the
perpendicular bisectors of all four wheels intersect
at a single point. In four wheel steering, this point
lies on a line between the front and rear wheels
running perpendicular to the fore-aft direction of the
base. This is illustrated in Figure 6. In two wheel
steering, the perpendicular bisectors of the front
steered wheels intersect at a point along the line
through the centers of the fixed rear wheels (Figure
5).
Outcomes Summary
A photograph showing a section of the
prototype steering linkage is shown in Figure 7.
A working platform that can demonstrate the
potential of the four-wheel drive configuration was
completed but the testing remains to be completed.
Recommended Future Research and Development
Future research and development should begin
by investigating the control issues concerning the
operation of a four wheel steered wheelchair. The use
of four wheel steering in the wheelchair application
introduces a dilemma for the control of that vehicle.
Optimum performance is likely attained when the
wheels can be left at arbitrary, but a known, steering
angle while the wheelchair is idle. Under these
conditions the driver knows which direction the chair
will initially go and there is no delay in initiating a
move. However, to make the direction of the wheels
known to the driver while the chair is at rest requires
the driver to observe the direction using a visual
inspection of the wheels or the direction information
must be provided using some other feedback
mechanism. Three options come to mind: 1) a visual
display on the controller panel; 2) tactile feedback
through the control stick using a rotation about either
the unused vertical axis or a rotation about the
steering axis; 3) no feedback at all. Although no
solution is ideal, a rotation of the stick seems more
desirable from the users perspective because it will
not require the driver to read a display, thereby
diverting his or her attention away from the
surrounding environment. The rotation option is
fronttypical pivotpoint
pivot point forminimumturning radius
Figure 6 - Wheel alignment for four wheel steering about asingle pivot point
Figure 7 - The complete linkage consists of two slidingmembers (A), four cam follower slots (B) cut into a flat plate(C), and two links (D) for each wheel.
13FINAL REPORT: 1993-1998
likely more complex and expensive to implement. The
third option, no feedback at all, will require the driver
to sense the wheel direction by sensing the direction
of travel once motion is initiated; this option is likely
to be problematic in confined spaces where the chair
is close to obstacles.
The other alternative for control of the vehicle is
to program the controller to self-center the wheels
each time the chair stops. This solution is also less
than ideal. In this configuration, there will be a delay
between the time when the user steers the wheels and
when the chair is able to travel in the desired
direction. If there is no direction feedback for the
wheels, the user is required to perform a visual
inspection of the wheel direction or sense the direction
after initiating a move by observing the direction of
travel.
Publications
Brienza, DM and Brubaker, CE. A four-wheel steering
mechanism for short wheelbase vehicles. Proceedings
RESNA Annual Conference, Pittsburgh, PA, June 1997
Brienza DM and Brubaker CE. A steering linkage for
short wheelbase vehicles:Design and evaluation in a
wheelchair power base. Journal of Rehabilitation Res &
Dev.1999;36(1)
US Patent No. 5,862,874.
14 RERC ON WHEELCHAIR TECHNOLOGY
TASK: PM-3 INPUT DEVICES AND CONTROL CONCEPTSInvestigators: David Brienza, Wonchul Nho, James Protho, Patricia Karg,
Jennifer Angelo and Kimberly Henry
Rationale
The interface between the wheelchair user and
the wheelchair itself is often the most critical
component of the powered wheelchair. Hand
operated joysticks with proportional control are now
the traditional method of interface for most
wheelchair users. Sip and puff control, head control,
chin control, single switch are further options for
those that are unable to access the joystick.
Goals
The objective of this task was to review existing
input and controller technology and explore technical
options for enhanced performance, reliability and
safety given current market needs and the evolving
national standards for microprocessor-based
wheelchair controllers.
Methods and Outcomes Summary
The results of a focus group meeting to identify
the most significant issues impacting input devices
and control concepts for powered mobility devices
held during the first funding period has been reported
(Brienza, et al, 1995).
A research and development plan consistent with
the needs identified by the focus group and
compatible with the goals and objectives of the RERC
was conducted. The long-term goal of this research
is to develop a control system that integrates
navigational and obstacle detection sensors into a
control system that assists the driver of a wheelchair
in both known, i.e., mapped, and unknown
environments. Potential applications of the system
include obstacle avoidance in known and unknown
environments, execution of predefined maneuvers
such as traversing through a doorway or following
along a wall, assisted navigation along predefined
paths through a known environment and as a driving
skills training device for powered wheelchair users.
Developments during the first project period
concentrated on the application of assisted obstacle
avoidance using a force feedback joystick. During the
second period the two control algorithms were
further developed.
Two philosophies have guided the design process:
1) ultimate control of the wheelchair must remain
with the driver and not with the control algorithm;
and, 2) mobility efficiency must be maximized.
Providing the user with the ability to apply the
decisive control input signals distinguishes this
wheelchair control system from that of an
autonomously guided vehicle. The driver remains in
control of the decision making element of the system
and at no time is an action initiated without allowing
the user to override the suggested action. Also, any
input action should result in a predictable response
from the system so that the user is not required to
decipher the control algorithm in order to accomplish
a desired task.
The object of the control system is to assist the
driver in negotiating obstacles as fast as possible and
with as little cognitive and physical effort as possible.
It is undesirable to slow down the wheelchair. This
would decrease efficiency or burden the driver with
excessive monitoring tasks, making the wheelchair
more difficult to drive. Instead our objective is to
influence the steering of the wheelchair using force
feedback from the active joystick. Note, however, that
the user may choose to counter the suggestions of
the control system by overcoming the joystick’s force
resistance.
Since the conceptual development of these control
modalities, this task concentrated on implementation
of a system for the evaluation of the concept.
An evaluation of a force feedback joystick for a
powered wheelchair was performed. The study aim
was to determine if the device enhanced the driving
performance of experienced wheelchair users. A
prototype device was constructed and used with a
virtual reality system for the evaluation phase of the
15FINAL REPORT: 1993-1998
study. The force feedback joystick is shown in Figure
8. Test subjects used the force feedback joystick as a
prototype to navigate a wheelchair through a virtual
environment with and without the force feedback
algorithm activated (Figure 9). According to the
position of the wheelchair in the virtual environment,
the force feedback algorithm changed the compliance
of the joystick making it more difficult to move the
joystick in the direction of an obstacle. The factors
that were used to determine the compliance of the
joystick were 1) the angle between the wheelchair
velocity vector and the displacement vector of the
closest obstacle, and 2) the speed of the wheelchair.
The subjects were experienced power wheelchair
users with marginal ability to control a wheelchair
using a conventional proportional joystick. Their
performance using the force feedback joystick was
measured using the time needed to complete a run
through the course and the number of collisions with
the obstacles. The test course is shown in Figure 10.
The results showed that one out of the five subjects
who participated in the study had fewer collisions
when the force feedback algorithm was activated
compared to their performance when the algorithm
was not activated.
Figure 8 - Picture of force feedback joystick
Figure 9 - Picture of a subject using the system
Figure 10 - Diagram of test course.
Publications
Brienza, D.M., Angelo, J.A., Henry, K. Consumer
participation in identifying research and development
priorities for power wheelchair input devices and
controllers. Assistive Technology, July 1995.
16 RERC ON WHEELCHAIR TECHNOLOGY
Rationale
Persons with limited motor abilities and multiple
technical needs are able to access assistive
technologies through either many individual
switches or integrated controllers. There are no
guidelines to assist clinicians or consumers in
identifying persons who will be successful users of
integrated controllers.
Goals
1. To determine criteria necessary for successful use
of integrated controls by persons with multiple
technology needs and complex physical
conditions.
2. To identify service delivery components which
support the recommendation and provision of
integrated controls.
Methods Summary
A survey for interviewing successful users of
integrated controls was developed in conjunction
with the Office of Research at the University of
Pittsburgh. Survey topics included, but were not
limited to, user characteristics, environmental factors,
amount and type of training and back up and
maintenance of systems. Respondents were located
through clinicians that worked in North American
institutions that were multidisciplinary and known
for their work in assistive technology. Thirty
clinicians were contacted and assisted in the
recruitment process. The survey was administered
over the telephone and results tabulated and
analyzed. A Likert type ranking system was used to
analyze survey results.
Outcomes Summary
Twenty-four people with severe physical
disabilities, who used integrated controls,
participated in the telephone survey. The survey
focused on their satisfaction with areas related to use
of an integrated control device. Respondents were
generally satisfied with their integrated control
devices. A moderate correlation coefficient was found
between gadget appeal and satisfaction with devices.
The sample was self-selected and voluntary.
Three areas were identified as leading to
satisfaction with integrated controls. One, the
introduction of the integrated controller gave the
respondents a method of accessing devices that, prior
to receiving the controller, they were unable to
operate. Second, some form of training took place.
Either the trial or error or trial and error plus a manual
were used for training in cases where persons were
satisfied with their integrated controllers. This
information might help clinicians select a training
method. Finally, persons who liked gadgets were
more likely to be satisfied with integrated controllers.
A second survey was completed with clinicians that
recommend integrated controls. Issues affecting their
recommendation of integrated controls included the
availability of technical support and the comfort of
the clinician with the technology.
Due to the small sample size and the fact that the
group was self-selected, the results must be
interpreted carefully and should not be generalized
to the population of persons using integrated control
devices. Further studies need to be conducted to
support or refute these findings. One group that may
be surveyed is the population that has abandoned
integrated control device to examine why the devices
were abandoned. Another area that should be
investigated is how these results differ when
surveying children. The device procurement,
receiving devices all at once or over time, the learning
curve, and type of training may be quite different
depending on the age and experiences of the
individual user. This survey demonstrated that
persons using integrated control devices were, in
general, satisfied with them.
TASK: PM-5 THE USE OF INTEGRATED CONTROLS BY
PERSONS WITH PHYSICAL DISABILITIES
Investigators: Jennifer Angelo and Elaine Trefler
17FINAL REPORT: 1993-1998
Recommended Future Research
Authors propose that a survey should be
conducted on the population that has abandoned
integrated controls. Another area that should be
investigated is how these results differ when
surveying children rather than adults. Finally,
training methods utilized with complex high
technology systems need to be investigated.
Publications
Angelo, J., Trefler, E. (1996). Surveying satisfaction of
integrated controls users. Proceeding of the RESNA
‘96 Annual Conference, Salt Lake City, UT, June 1996:
212-214.
Trefler, E and Angelo, J. Surveying Users of
Integrated Controls - A Pilot Study. Proceedings,
ARATA. Adelaide, Australia, October 1995: 17-19.
Angelo J and Trefler E, (1998), Satisfaction of Persons
Using Integrated Controls, Assistive Technology, 10.2.
77-83.
18 RERC ON WHEELCHAIR TECHNOLOGY
Rationale
Very few powered wheelchairs have been
optimized for activities conducted in tight indoor
environments. Reaching up and down, transferring
and maneuvering in confined spaces are examples
of these activities. Many older persons with
disabilities have need for such mobility products, but
will often reject the notion if it makes a statement
about their disability. Aesthetics is an important
component to acceptance and, therefore, it was given
high priority in this task.
Goals
1. To provide increased indoor powered mobility
options for consumers of all ages and disabilities
with emphasis on environments of older persons.
2. Refine commercially promising designs and
facilitate transfer to the marketplace.
Methods Summary
The PM2 Advanced Mechanisms task addressed
the wheelchair steering problem by using a
mathematically designed cam and linkage steering
arrangement. This task addressed the need for
increased indoor maneuverability by using two
software-controlled servo-steering motors to control
the position of the two front drive motors. A
prototype, termed the PM6-MKI was developed
which also featured a novel tiller-type joystick control.
The software algorithm compensates for the
difference in turning radius of the two front driving
wheels and thereby minimizes any wheel scrubbing
effect. (Figure 11). The front wheel drive motors used
in the prototype were Fracmo, Model: M453-W30,
previously developed by Legal and Hobson. First
stage comparative maneuverability testing was done
using existing powered wheelchairs typically used
indoors as the benchmark.
A second design, the MKII, which grew out of
TASK: PM-6 NEW CONCEPTS IN
POWERED INDOOR MOBILITYInvestigators: Douglas Hobson, Linda van Roosmalen
Collaborators: Jules Legal, Steve Stadelmeier
our relationship with the students and faculty in the
Design Department at Carnegie Mellon University,
is shown in Figure 12. The Quality Function
Deployment (QFD) [Jacques et al., 1994; Logan &
Radcliffe, 1997] tool was used to establish the design
criteria. This prototype addresses the need for
improved esthetics and self-adjustability of seat
height and angulation. Re-cycled motor drives
combined with a standard controller were used to
power the prototype. Two linear actuators control the
height and inclination of the seat.
The task plan called for the combining of the best
features of each prototype into a final demonstration
product. The full implementation of this plan was
dependent on the availability of a new motor drive
system, which was the focus of MK I prototype and
task PM-1d. In spite of several efforts at working
directly with motor drive manufacturers, we were
unsuccessful in convincing a company to invest
resources in a newly configured motor drive system.
Illustrations of the MK I and MK II designs follow.
Figure 11 – PM6-MK I Evaluation Prototype
TILLER-TYPE CONTROL
POWEREDSTEERING
STEERINGMOTOR
19FINAL REPORT: 1993-1998
Concept illustration based on QFD criteria Working Prototype
Figure 12 - MK II Prototype
Figure 13 - Corridor Figure 14 - Bathroom
20 RERC ON WHEELCHAIR TECHNOLOGY
Outcomes Summary
a) Laboratory Feasibility Testing of the PM-6 (MK
I) Prototype
The purpose of the feasibility test was to compare
the maneuverability of the MK I prototype to that of
production wheelchairs designed for similar usage.
Two production wheelchairs, Quickie P190 and the
E&J Tempest, were selected for the tests. The tests
consisted of running the three wheelchairs through
three typical environmental spaces setup as a
laboratory test course. Each space was laid out
according to the dimensions of the Uniform Federal
Accessibility Standards.
The course setup consisted of the following three
spaces as shown in figures 13-16 below. The
dimensions of the test spaces are as follows:
Corridor: w=91.7 cm; Bathroom: w x d=152.3 x
142 cm; Elevator: w x d= 171.2 x 129.3 cm
Walls for each space were fabricated from
replaceable 3/4” thick polystyrene foam sheets,
which showed damage marks each time they were
contacted by a wheelchair.
Test Method
The MK I, Quickie P190 and the Tempest
wheelchairs were randomly assigned to 4 test
subjects, all non-experienced wheelchair users. The
subjects were all given the same time to become
familiar with the standardized test course. They were
then asked to maneuver through the test course, twice
with each wheelchair.
Time was measured for each wheelchair to
maneuver through each space. The time started when
the front feet of the test wheelchair passed the space
threshold line. The time was stopped when the
wheelchair exited past the space threshold line. Also,
within each space the number of hits with the course
“wall” was recorded.
The first space, the corridor, was entered in a
forward direction. The subject had to first steer the
Figure 15 - Elevator
Figure 16 - Overview of the complete course layout. Thelines indicate the required maneuvers.
Wheelchair Powered by Front wheel type Footprint
MK I Powered front wheels Steered powered wheels 80 x56 cm
Quickie P190 Powered rear wheels Swivel caster 107 x 61 cm
E&J Tempest Powered rear wheels Swivel caster 94 x 65 cm
TEST WHEELCHAIR DATA
21FINAL REPORT: 1993-1998
wheelchair into the right corridor and proceed until
they could touch a designated point on the wall with
their hands. They then backed down the corridor until
they could turn right and exit through the entrance
corridor.
The bathroom space had to be entered in a
forward direction. An object on the simulated vanity
was touched. The subject then backed out of the
bathroom.
The elevator space was approached in a forward
direction. The subject then turned 180 degrees and
touched the simulated control buttons for the elevator.
The subject then exited the elevator forward facing.
Results
The test results were analyzed in such a way that
the maximum speed of each wheelchair did not
influence the outcome of the test. The sample results
of the tests are shown in the following graphs. The
first two graphs are for a single subject; the last two
are the averages for all subjects.
Figure 17 - Average test time of subject #1 per space for the three test wheelchairs
The graphs indicate that in most cases the PM6 -
MK I wheelchair resulted in the shortest test time and
the least number of inadvertent walls impacts. Little
difference was seen in the time needed for the
washroom test. The reason for this may be that the
overall maneuvering requirements of the space were
not extensive. Whereas, in the corridor test, most
subjects took substantially longer to maneuver with
the Tempest and the Quickie wheelchairs than with
the MK I wheelchair. Finally, the elevator test was a
time consuming task for all three wheelchairs. In
terms of wall impacts, the graphs indicate that the
PM6-MKI wheelchair clearly performed better then
the other two test wheelchairs.
Average Test Time 1
0
10
20
30
40
50
60
Corridor Elevator Washroom
Average time (sec)
PM-6: 4.48m2Tempest: 6.11m2Quickie P190: 6.53m2
22 RERC ON WHEELCHAIR TECHNOLOGY
Figure 18 - Average number of wall impacts by subject #1 for each test wheelchair
Figure 19 - Average number of wall impacts for all subjects for the three wheelchairs/spaces
Amount of Hits per W/C 1
0
1
2
3
4
5
6
7
8
9
1 2 3
Averageamount
of hits (n)
PM-6: 4.48m2Tempest: 6.11m2Quickie P190: 6.53m2
Average Number of Hits
0
2
4
6
8
10
12
14
Numberof hits (n)
PM-6: 4.48m2 0.5 2.5 1
Tempest: 6.11m2 12.5 12 4
Quickie P190: 6.53m2 11.5 8.5 2
Corridor Elevator Washroom
23FINAL REPORT: 1993-1998
Figure 20 - Average test time for all subjects for the three wheelchairs/spaces
Discussion
All wheelchairs used in the tests had different
‘footprints’, the MK I being the smallest. Therefore,
direct comparisons and any conclusions from the
results must be done with caution. For example,
reduction in the footprint size of the production
wheelchairs to that equal to the MK I wheelchair
would most likely improve their wall impact
performance. Also, the difference in maneuverability
times could be effected by the larger footprint size of
the production wheelchairs and not be totally due to
the enhanced maneuverability of the MK I prototype.
The Tempest and Quickie wheelchairs have front
swivel casters, which makes it impossible to
maneuver backwards from a forward maneuver
without first causing a lateral ‘shift’ of the front end
of the wheelchair. This was, in some cases, the reason
for higher number of wall impacts of the production
wheelchairs. Whereas, the MK I wheelchair, having
powered steering of the front wheels, does not exhibit
lateral shifting when reversing course.
Finally, because of the small size of the test
sample, no statistical analysis was attempted.
Therefore, it is only an observational conclusion that
can be drawn from this simplified feasibility test.
As mentioned, the MK I design also features a
uniquely designed tiller-type joystick. The idea is that
most elderly people will intuitively relate better to
tiller control (side to side movement to steer up and
down for reverse and forward, respectively). Also, the
direction of the tiller could be coupled electronically
to the direction of the steered wheels, so at start-up
there would be no directional surprises. Although this
joystick design worked well during the tests, no
comparative tests with the conventional joystick were
possible.
b) Development of the MK II Design
In brief, the purpose of the MK II design was to
explore the following criteria for an indoor wheelchair
that would provide:
• an alternative to the scooter for indoor/home use,
• an economic way to give elderly people mobility
in institutional settings,
• an alternative for the indoor/outdoor for home
to office use,
• an alternative for ADA accessibility into tight
workspaces, offices,
Average Test Time
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
Time (sec)
PM-6: 4.48m2 21.0 15.0 10.2
Tempest: 6.11m2 37.4 19.2 15.3
Quickie P190: 6.53m2 40.1 20.7 9.4
Corridor Elevator Washroom
24 RERC ON WHEELCHAIR TECHNOLOGY
• a better way to vary the sitting height of a person
in a W/C, and
• a more esthetic and less stigmatizing way of
providing powered mobility.
Focus groups, user surveys and the Quality
Function Deployment (QFD) tools and the MK I
experience were used to explore questions and solicit
concepts leading to a list of weighted design criteria.
A sample questionnaire containing comments from
wheelchair users can be reviewed in Appendix A. The
summary results of the QFD analysis are also
contained in Appendix A. The MK II prototype shown
in figures 21-22 resulted from these intensive
planning efforts. The working prototype embodies
the following key features:
• a nontraditional frame and elevating/tilting seat
system,
• a ergonomically designed seat with swing up
armrests for ease of transfer,
• powered front wheels, castered rear wheels
allowing increased maneuverability in tight
indoor spaces. (Steered front wheels were
planned but suitable units were not possible to
obtain for the prototype construction),
• miniature integrated joystick control,
interchangeable between left and right armrests,
and
• a non wheelchair-like appearance intended to
minimize the stigma of disability.
The Mark II design was featured at the 1998
RESNA Conference exhibit. Interest was
demonstrated by clinicians, wheelchair users and two
prospective wheelchair manufacturers. Below are
several photos showing some of the features of the
MK II design.
Recommended Future Development
Given that both demonstration outcomes were
basically positive, this development now requires
significant resources to integrate the best of the
demonstrated MK I & II features, complete with a
newly developed motor drive system. It will require
the formation of a partnership between the
developers, and, at least, a committed wheelchair
Figure 21 - Seat raises and tilts to aid in standing. Foot rests
drops to floor.
Figure 22 - Arm rests flip back to aid in transfer and work
place access .
25FINAL REPORT: 1993-1998
manufacturer and motor/drive developer-supplier
to transition the development towards commercial
availability. The investigators have made plans for
the formation of such a partnership and an SBIR
submission is under preparation to help finance the
venture. Assuming success with the SBIR submission,
the plan calls for the development of an integrated
MK III design. The MK III will then be subjected to
more rigorous laboratory and user testing as part of
its Phase I feasible evaluation.
Publications (in preparation)
References
Jacques GE, Ryan S, Naumann S, Milner M, Cleghorn WL
Application of Quality Function Deployment in
Rehabilitation Engineering, IEEE Transactions on
Rehabilitation Engineering, Vol. 2, No. 3, September 1994.
Logan GD, Radcliffe DF Potential for use of quality matrix
technique in rehabilitation engineering. IEEE Transactions
on Rehabilitation Engineering, Vol. 5, No. 1, March 1997.
Brown PG, QFD: Echoing the voice of the customer, AT&T
Technical Journal, March/April, 1991, pp. 18-32.
Hauser JR, Clausing D The house of quality, The Product
Development Challenge, Harvard Business Review Book,
eds. Kim B. Clark and Steven C. Wheelwright, pp. 299-
315, 1995.
26 RERC ON WHEELCHAIR TECHNOLOGY
TASK: PM-7 POWERED MOBILITY SIMULATORInvestigators: Douglas Hobson, Nigel Shapcott, Mark Schmeler
Collaborators: Robert Lang, Jules Legal
Rationale
Evaluation for powered mobility can be a difficult
and time consuming process for both service
providers and wheelchair users. The decision to
recommend or purchase a powered wheelchair must
be done carefully and with maximum consumer
involvement as the costs are often high and the
mistakes are difficult to rectify after the fact. For
individuals with severe disabilities, the selection
process can often involve trials with different types
of input controls in an effort to determine if powered
mobility is even a viable option. For others who have
been long time manual wheelchair users, manual
propulsion may become increasingly more difficult
as a result of progressive disability or older age. A
powered wheelchair simulator is a multi-purpose tool
that allows consumers and clinicians to experiment
with powered mobility options at a relatively low cost
in an effort to make informed decisions prior to the
purchasing process. It allows a person in their manual
wheelchair, in their typical seated posture, to
experience the sensation of being in a powered
wheelchair. The concept is based on having a
powered platform or simulator onto which a person
can wheel their manual wheelchair. Controls can be
readily selected and positioned to meet the individual
needs of the user. Assuming the trial is positive, the
clinician, working closely with the user and assistive
technology supplier, can then more confidently
formulate the specifications for the definitive
powered wheelchair. This approach can be a
significant improvement over the typical trial and
error approach, as well as reduce the chances of
prescription error and ultimate disappointment by
the user. Research work conducted by Mark Schmeler
and Nigel Shapcott, while at the University of Buffalo,
indicates that the sensation experienced by users
while on the simulator closely parallels the motor/
perceptual sensations experienced in an actual
powered wheelchair (Schmeler, ’95).
Methods Summary
A first generation prototype simulator was
constructed during the latter part of Year II.
Evaluation of the first generation prototype was
performed in the University of Pittsburgh Medical
Center’s Center for Assistive Technology (CAT). A
local assistive technology supplier was invited to
participate in the prototype implementation and
evaluation. The results of this interaction were
positive, including suggestions for MK-II design
improvements. A mechanical designer (Jules Legal)
was added to the team. An unsuccessful STTR
proposal was prepared and submitted to NIH/
NCMRR in partnership with two local firms in Yr.
III. Year IV focused on continued refinement and
testing of the MK-II design with consumers in the
CAT. Based on the positive local experiences, a
second, revised STTR grant proposal was submitted.
It was not successful. The CAT also produced several
units for use by other clinical facilities.
Figure 23—Powered Mobility Simulator prototype based on the
Suny-Buffalo design.
27FINAL REPORT: 1993-1998
Outcomes Summary
Transfer of this development to the marketplace
was dependent on a commercial partnership and the
receipt of technology transfer support from external
sources. As indicated above two attempts at securing
the necessary federal support were unsuccessful.
Several reviewers questioned the viability of such a
product given its limited application and therefore
numbers that can be potentially sold. This may
possibly be the case. However, we were gratified that
Mark Bresler, [Bresler, ML, 1990], one of the early
proponents of the wheelchair simulator concept,
exhibited a new prototype at the 1998 RESNA
conference. Hopefully he has captured the interest of
a commercial entity that will make this “orphan”
development available to those clinicians in most
urgent need.
Publications
Schmeler, M.R. Performance Validation of a Powered
Wheelchair Mobility Simulator, Proceedings of theEleventh International Seating Symposium, Pittsburgh,
PA, February 1995
References
Bresler, MI Turtle trainer: A way to evaluate power
mobility readiness. Proceedings of the Thirteen AnnualRESNA Conference, 1990
28 RERC ON WHEELCHAIR TECHNOLOGY
Rationale
The purpose of this project was to design and
develop a novel wheelchair with a unique
combination of features. This wheelchair design was
intended to address a market need for a wheelchair
capable of folding compactly for stowage (e.g.,
overhead compartments during commercial air
travel), accessing narrow passageways and other
areas requiring a compact profile and footprint, and
providing a high degree of maneuverability. Our
intent was to design an “Enhanced Access
Wheelchair” to achieve these capabilities without
sacrificing the performance characteristics essential
for everyday use (Figure 24). We also attempted to
determine the feasibility of incorporating fiber
reinforced material technology.
TASK: MM-1 STRUCTURAL IMPROVEMENTS TO
MANUAL WHEELCHAIRSInvestigators: Clifford Brubaker, David Brienza
Collaborators: Phil Ulerich and Catherine Palmer, Westinghouse Corp.
Figure 24 - Enhanced Access Wheelchair.
Goals
The goals of this project, as originally proposed,
were to design, fabricate and evaluate an aesthetically
pleasing, general-purpose wheelchair that could be
easily stowed, manipulated and maneuvered through
narrow corridors. The conceptual design was
proposed to meet this objective by providing several
important features:
• Compact folding frame;
• Light weight (using composite materials);
• Three position (anti-tip, rear support and folded)
auxiliary wheels; and
• Attractively shaped solid side frame members
that allow for subtle incorporation of mechanisms
like brakes, releases and structural members.
Three prototype wheelchairs were designed,
fabricated and tested in the course of this project. The
final design incorporated side frame members made
from inexpensive, molded thermoset materials. Other
frame components were eventually machined
individually from aluminum stock due to difficulties
with machined composite parts. It was (and is) our
expectation that these parts could be manufactured
more efficiently in production models. The
dimensions of the folded frame are 5.5 inches wide
by 21 inches deep by 12 inches tall with the footrest
mounts and main wheels removed and not including
the back rest. The wheelchair is shown folded in
Figures 25 and 26. Auxiliary wheels are included to
allow passage through openings as narrow as 18
inches (overall width is defined explicitly by the seat
width) when the main wheels are removed (Figure
27). The development effort has focused on the
incorporation of these novel features and
manufacturing processes. Weight reduction will be
an objective for subsequent design iteration.
An important secondary goal of this project was
to demonstrate alternative materials and
manufacturing techniques for the production of
wheelchairs. Our experience with this option is
summarized in the following section of this report.
29FINAL REPORT: 1993-1998
Figure 25. - Side view of folded wheelchair.
Methods
A CAD design for the prototype wheelchair was
executed using CADKEY. This design was exported
to a more sophisticated CAD system at Westinghouse
Corp. Science and Technology Center where the
design was further refined. A structural analysis
using ANSYS, a finite element analysis program, was
conducted to determine the necessary material
strengths for the different parts. Stress analyses were
performed on individual components and for an
articulated model of the prospective prototype. Upon
completion of the design and computer simulation
phases, the project proceeded to the development of
the physical prototype. Thermoset materials were
considered as a low-cost production option for
wheelchair structures.
Inexpensive, molded thermoset materials offer
several advantages for use as low cost wheelchair
structures. Two major disadvantages are
manufacturers’ lack of experience with thermoset
molding and the high initial cost of molds. Both of
these problems were considered in this project.
The structural elements of the wheelchair were
designed as compression molded, glass filled
polyester components. One reason for this selection
is the very low cost of this material. It is used
commonly in industry for electrically insulated
structural parts. Since it is an engineered plastic, an
entire range of material strengths, weights and costs
are available. This allows for trade-off between
weight and cost in the design and manufacture of
wheelchairs. The basic design and geometry of this
wheelchair was defined substantially by the novel
folding mechanism of the chair and by common
structural requirements for wheelchairs.
A few iterations of weight reduction analysis were
done on the parts to save some material. Considerably
more refinement is possible. The chair was modeled
as plate elements and loaded with a 200 kg dummy
at 3 g’s. Consideration was given to both the
maximum von Mises stress and the maximum
deflection. Acceptable deflection was based only on
assumed aesthetic perceptions for the prototype
development. The stress limit was determined from
isotropic treatment of the maximum allowable tensile
stress.
The high cost of mold fabrication precluded mold
development for parts other than the side frame. Parts
were initially machined from sheet stock. This
decision was made with the knowledge that
machined composite parts typically have structural
strengths on the order of 40% less than comparable
Figure 26 - Folded (front).
30 RERC ON WHEELCHAIR TECHNOLOGY
molded parts. This loss of strength is well
documented and results from surface cracks and
defects left from milling the smooth, fiber free
surfaces. The use of machined composite parts proved
not to be a viable solution in subsequent testing. The
parts (other than the side panels) were subsequently
machined from aluminum sheet and bar stock.
wire is fed into the spray head and electrically melted.
Repeated layers of sprayed metal were applied until
a shell was created from 1/8 to 1/4 inch thick. This
shell was backed with an aluminum-filled epoxy to
provide strength and stiffness and placed into a cold
rolled steel frame about 1/2 inch thick. This process
was repeated to form the other half of the mold.
Unfortunately, the mold was incorrectly developed
as a conventional injection mold, rather than as a
compression mold. For injection molding the mold
is parted at the midline with two symmetrical (in this
instance) halves that are held in opposition while
material is injected. In contrast, a compression mold
has a “force” component and a “cavity” component
as the two “halves.” Without a force and cavity, it
was difficult to assure that sufficient material would
be incorporated into the mold to fill the part. After
six attempts the proper charge of bulk molded
material to fill the part was determined. After the
third piece was molded, the ejector system failed and
the molder was forced to pry subsequent pieces out
of the mold using hand tools. This was difficult, as
the mold must be stabilized at 350 degrees before the
molding process can begin. The failure required a
modification of the mold. It became necessary to
machine away extra material. This ultimately
weakened the parts.
The molding technology chosen for this project
is based on spray metal tooling. This technique for
mold making takes about a month and costs less than
$8,000. This process is rather new and has seldom
been used on compression molded parts of this size.
Only 20 to 150 parts would be expected from this tool.
By contrast, standard mold construction (using steel)
for comparable sized parts would require 4 to 6
months to complete at a cost on the order of $70,000.
These steel molds could be used to produce 500,000
to 5,000,000 parts. Standard aluminum molds are less
expensive ($45,000), quicker to machine
(approximately 3 months), and would be suitable for
producing 5,000 to 25,000 parts. The project provided
an opportunity to consider the efficacy of
compression molded parts at modest cost.
The mold was fabricated over the course of 8
weeks and was received at Penn Compression near
Pittsburgh, PA. The mold was made from a wood
model of the final part, which was placed in an inert
bed up to the mold parting line. An electric spray head
was used to sputter-coat thin layers of a zinc-
aluminum alloy onto the pattern. Zinc-aluminum
Figure 27 - Side view with main wheels removed.
Figure 28 - Narrow access.
31FINAL REPORT: 1993-1998
The molded side-frame had four “through
holes,”including a 1" diameter main axle hole and
three 1/4 inch diameter holes to stop the auxiliary
wheel in its various positions. Of the parts produced,
five did not fill completely and several others were
broken while being ejected from the tool. In the end,
six acceptable parts were made, allowing for the
assembly of three prototypes.
Prototype assembly
Fabrication of an initial prototype resulted in the
discovery of weaknesses in the original design. As a
result of the initial fabrication phase, a substantial
number of the components were redesigned with the
goal of increasing the structural integrity of the
wheelchair frame. The modified designs were used
to fabricate two additional prototypes, which were
evaluated using applicable ISO standard test
procedures. The modified version is shown in Figure
29 and 30.
fatigue strength). The chair passed all static and
impact strength tests with the exception of armrest
upward weight bearing. The armrest upward force
test is not applicable to our design since the armrests
were designed to release with upward force. The chair
successfully completed 200,000 cycles on the two-
drum fatigue strength test without failure, but failed
after 2055 cycles of the curb drop test. This failure
was a fracture of the side frame where the footrest
and front caster wheels are attached. Prior to the
testing, we observed cracks in the frame resulting
from a poor fit between the molded side frame and
the footrest/caster wheel mount. It will be necessary
to address this area of structural weakness in the
design of future prototypes using molded
components.
Consumer Evaluation of the Prototype
Initial evaluation was provided by an
experienced wheelchair user and resulted in several
comments and suggestions:
• The concept of the design is attractive. The ability
to remove the rear wheels and use 8 inch auxiliary
wheels to roll down an airplane aisle or in a small
rest room would be useful. (Figure 28)
• The ability of the chair to fold and break-down
into small components makes it attractive for
storing in overhead compartments of aircraft or
in compact automobiles.
• The folding mechanism is awkward and
cumbersome. The wheelchair can become difficult
to fold if the central pin loses alignment with the
cross-braces. The dovetail joints bind and are
prone to jamming from dust and dirt.
• The wheelchair is much too heavy. The materials
need to be changed and the overall design
lightened.
• The wheelchair is too tall and the leg rests are
positioned too far forward.
• The auxiliary wheels do not perform adequately
as anti-tip devices and are cumbersome to use.
• The wheelchair and center of gravity are not
adequately adjustable.
• The backrest folding mechanism is bulky and
does not provide adequate lateral stiffness.
Figure 29 - Assembled Prototype.
ISO Test Evaluation of the prototype
The first of the modified prototypes was tested
according to ISO 7176-8 (Wheelchairs - Part 8:
Requirements and test methods for static, impact and
32 RERC ON WHEELCHAIR TECHNOLOGY
• The chair has multiple pinch points that need to
be eliminated
• The wheelchair provides a proof-of-concept and
would require additional refinement prior to
being acceptable to consumers.
wheelchair that allows access to narrow corridors and
is rigid and durable enough for everyday use is now
several years old, there is still considerable need for
such a product by many wheelchair users. As a result,
we feel that the market potential for a wheelchair with
these features is still significant. This initial funding
provided the basis to take the most critical step in the
research and development process: from conceptual
design to full-scale working prototype. There are still
several important engineering problems to solve
before the eventual development of a commercial
product; however, we have successfully
demonstrated the feasibility of producing the
wheelchair for enhanced access. Although it was not
our primary objective, we have also shown the
possibility of using parts manufactured with
inexpensive techniques and materials.
External Evaluation
A more thorough demonstration and evaluation
was conducted by the RERC on Technology Transfer
at SUNY Buffalo. The Enhanced Access Wheelchair
was evaluated by three focus groups of 30 consumers
who had used a manual wheelchair for a minimum
of five years. Some general results from comparisons
with existing commercial products were particularly
encouraging:
1. 55% of the consumer participants preferred the
prototype to existing products.
2. Consumers were willing to pay up to $200 more
for the features incorporated in the Enhanced
Access Prototype.
3. Consumers increased the additional amount they
would pay for the prototype features to $370
(mean) after viewing the features of a competing,
production model wheelchair.
4. Among features valued by the consumers were
the folding mechanism, the folded size, and the
3-position deployment of the auxiliary wheels, the
“solid” seat, and the aesthetics of the side-frame.
Disadvantages identified included the
imprecision and “awkwardness” of the
mechanisms, the overall weight, and the lack of
tie-down points. Suggestions were generally on
ways to improve the mechanisms and decrease
Figure 30 - Front view of assembly
Several of these problems have already been
addressed. For example, the seat was redesigned to
eliminate the possibility of pinching while it is being
opened. The backrest support brackets were
redesigned since this initial evaluation was made.
Amelioration of all other shortcomings is being
considered. The design will require further iteration
to become viable for commercial development.
Outcomes Summary
Our initial impression of the prototype relative
to its performance is positive. The solid seat together
with the cross braces and side frame members form
a support structure that feels significantly more rigid
than typical “X” cross brace frame, folding
wheelchairs. Even with the large main wheels
removed for narrow access, the wheelchair was
sturdy and stable. In informal trials in our laboratory,
varying users have found the chair’s performance to
exceed their expectations for a folding frame
wheelchair. A formal beta test program will be
developed as the next stage of development.
Although the concept of a compactly folding
33FINAL REPORT: 1993-1998
the weight. One of the strongest preferences for
the prototype over production folding chairs was
the folding mechanism. It was perceived to be
more stable and to allowed more compact folding.
Recommendations for Future Development
The project has progressed to the point of
successful demonstration of several valuable features
of a manual wheelchair. We believe that the
evaluation information is sufficiently positive to
warrant further development. Initial plans for a
second generation prototype have been completed.
We believe that it will be necessary to produce a metal
frame model to gain the interest of current
manufacturers. If we can obtain additional funding
for this project we shall proceed with development
of an all metal prototype in which we shall refine the
mechanisms and reduce the weight of the wheelchair
as suggested by the consumer panels.
Publications and Technical Reports
Brienza, DM, CE Brubaker (1996) Design and Development
of a Wheelchair for Enhanced Access, RESNA Proceedings,
16:250-252.
Brubaker CE, Brienza DM, Ulerich P “Design and
Development of a Wheelchair for Enhanced Access,” Final
Progress Report, SCRF Grant #1218, Paralyzed Veterans
of America, November 10, 1995.
Ulerich P, Palmer, K, Stampahar,M, Brubaker, CE “Design
and Development of a Wheelchair for Enhanced Access,”
First Annual Report to the Paralyzed Veterans of America
Spinal Cord Research Foundation, Grant #1218-01, March
16, 1994.
34 RERC ON WHEELCHAIR TECHNOLOGY
TASK: WP-1 CONSUMER RESPONSIVE MOBILITY
PRESCRIPTION PROCESS
Investigators: Elaine Trefler, Heather Rushmore
Rationale
Consumers who use manual wheelchairs have
expressed the view that their first wheelchair did
not meet their personal needs. The purpose of this
study is to develop a consumer responsive wheel-
chair prescription process for first time wheelchair
users who are functioning as paraplegics.
Over the past several years, consumer-respon-
sive services have become the highly studied
means of providing assistive technology and
rehabilitation services. In the past, consumers were
not given ample choices nor were they often asked
to contribute to the decision making process.
Often, all decisions were, and at times still are
today, made by the medical/therapy team. Due to
the lack of involvement by the consumer, he/she is
often dissatisfied with the assistive technology
received.
Goals
1. To determine the components of a service
delivery process that support consumer satis-
faction both with the process and the product
during the provision of their first wheelchair.
2. Propose enhancements to the service delivery
model based on the findings.
Methods Summary
The following steps were taken to address the
above goals:
1. Develop an interview instrument to determine
consumer satisfaction with the prescription
process for a first time wheelchair user, admin-
ister it to at least one individual to obtain input
into the areas that need refinement and gather
feedback to develop discussion areas and
questions for a focus group.
2. Form a focus group to gather ideas on ways to
improve the wheelchair prescription process.
Use input from the focus group to further refine
the interview instrument.
3. Identify and interview 30 consumers (15 who
received services from a multidisciplinary
clinical setting and 15 from a non-
multidisciplinary setting) with the interview
instrument developed to determine consumer
satisfaction with service delivery and wheel-
chair technology.
4. Review current practice based on information
and data collected from the focus group and
interviews.
5. Develop and propose enhancement to the
service delivery model based on the data
collected in steps 1-4.
Outcomes Summary
A focus group of five expert wheelchair users
was assembled to generate ideas on improving
current prescription processes. The group
brainstormed 31 ideas and ranked the top six ideas,
which were:
1. focus on the person;
2. consumer testing of different wheelchairs;
3. education on different wheelchairs for different
activities;
4. evaluation of the consumers home;
5. wheelchair user as a team member; and
6. peer counselor/mentor as part of the team.
Publications were prepared for both consumer
and professional publications, which summarized
the results of the focus group. The main themes
were that consumers wanted to be involved as full
partners in the decision making, to be able to try
different options in their own environment and
access to advice from other wheelchair users.
35FINAL REPORT: 1993-1998
Recommended Future Research
Based on the above experience we recommend
the following areas for further investigation:
1. Investigate effectiveness of education and
training methods for consumers. Document
consumers’ perceptions of training practices
and determine compatibility with active prac-
tice.
2. Compare consumer’s first prescription process
to their most recent to determine features and
satisfaction level.
3. Investigate the possible differences between
satisfaction and adjustment levels of individu-
als with acquired and congenital disabilities
and how this might relate to components of the
service delivery process.
Publications
Trefler, E and Rushmore, H. A consumer responsive
mobility prescription process: The summary of a focus
group. Team Rehab Report, June 1997, 41.43.
Trefler, E, Fitzgerald, S, and Rushmore, H. Manual
Wheelchair Prescription Process: Consumer Satisfaction
with Multidisciplinary and Non Multidisciplinary
Approaches, In revision.
36 RERC ON WHEELCHAIR TECHNOLOGY
TASK: WP-2 WHEELCHAIR PRESCRIPTION SOFTWARE
PROJECT (WPSP)Investigator: Nigel Shapcott
Rationale
Current wheelchair users and prescribers (OT, PT
and RTS students are the target population) have a
large and increasing selection of wheelchairs to
choose from, each having a variety of accessories that
customize the wheelchair to individual need. Thus,
the goal is to provide users the opportunity to
participate in the selection of the wheelchair that is
closest to being ideal for their needs.
Information overload caused by the significant
number of companies making wheelchairs, which
come in a variety of models with many configurable
options for each, leads to a large quantity of
information that has to be searched in order to make
appropriate selections. Information continually
changes as new models, options and companies enter
the market. Added to this is the fact that the
information between different manufacturers may be
difficult to compare because the wheelchair standards
testing information is not readily available.
Incorrect prescription or purchase of wheelchairs,
particularly among first time inexperienced
wheelchair users, is common among individuals with
spinal cord injury and other diagnoses where needs
change over time. There is a lack of training
opportunities that teach and inform prescribers on
the strategies of wheelchair prescription, taking
account of physical needs, functional environment,
funding and other issues, and relating these to the
priorities of a particular individual.
This collaborative project, to develop wheelchair
prescription software, has been funded mainly
through the Department of Veterans Affairs,
Rehabilitation Research and Development Service
(VA RR&D) as a component of the Computer Aided
Wheelchair Prescription System (CAWPS).
Goals
1. Develop a computer program that provides an
effective, easy to use wheelchair prescription
teaching aid.
2. Provide easy access to expert prescription
methodologies.
3. Commercialize the software in order to provide
a mechanism for widespread availability at
reasonable cost.
Methods Summary
An interactive computer based wheelchair
prescription system, using expert system
methodologies, has been developed. As part of this
development process, internal evaluation and
interactive evaluations were carried out using known
case studies.
Outcomes Summary
1) The software structure was stabilized January
1997. Educational features include:
• Quicktime videos to show different wheelchair
types and activities to educate and raise
expectations about what may be reasonable
achievements in education, work, leisure and
ADL activities.
• Graphics to explain dimensional information.
• Incorporation of a publication on wheelchair
selection as resource material (text and graphics).
(Axelson et al, 1994)
• Each question has an accompanying explanation
which can be accessed by a simple mouse click
(“Why Button”).
• Each feature of the final generic wheelchair can
be examined (simply by a ‘click’) to determine
which questions (and accompanying answers)
were factors in the selection of that feature.
37FINAL REPORT: 1993-1998
2) A demonstration version is available at:
ftp.pitt.edu/users/s/g/sgarand.
3) Logic developed has been largely completed
and is currently in the testing and editing phase.
4) Formal testing was not carried out in order to
secure funds and protect confidentiality pending
completion of negotiations with a potential
commercial partner (see below). The input from
informal testing has been very positive for this
educational version.
5) The project has been successful in attracting
commercial interest. An agreement, through the VA
RR&D Technology Transfer Section with a major
health care company who expressed interest in
commercializing CAWPS, failed. The company had
intended to further develop CAWPS and make it
widely available through Intranet and Internet links
as well as in a stand alone format. Task WP-2, WPSP
was planned to be released as a low cost (possibly
free) version of the main CAWPS program as part of
the commercialization plans. In December 1998,
negotiations ceased.
6) Plans are now under way to obtain funding
for further testing.
Individuals interested in obtaining a
demonstration version of CAWS should contact Nigel
Shapcott preferably by e-mail at Shapcott@pitt.edu
or through the RERC at 412-647-1273.
Recommended Future Development
1. Investigate the use of CAWPS as an
educational tool.
2. Investigate the use of CAWPS as an clinical
tool.
Publications
Shapcott, N and Garand, S. Computer-Aided
Wheelchair Prescription System, paper submitted to
Canadian Seating Symposium, Toronto, Canada, Sept
1996.
Shapcott, N and Albright, S. Computer-Aided
Wheelchair Prescription System, paper submitted to
Pittsburgh International Seating Symposium,
Pittsburgh, PA February 1997.
Reference
Axelson P, Minkel J, Chesney D. Guide to wheelchair
selection: How to use the ANSI/RESNA wheelchair
standards to buy a wheelchair, PVA 1994
38 RERC ON WHEELCHAIR TECHNOLOGY
TASK: STD-1 PARTICIPATION IN THE DEVELOPMENT OF
WHEELCHAIR STANDARDS
Investigator: Rory A. Cooper
Rationale
Development and application of performance
standards is perhaps the most productive activity in
terms of affecting the improvement to the quality of
wheelchair products for the largest number of users.
However, standards development requires research
and testing in order to validate the test procedures
prior to their acceptance in national and international
standards. Standardized disclosure of test and
measurement data in presale brochures is a means
by which consumers can accurately compare
products prior to purchase commitment.
Goals
1. To participate in the development and revision
of wheelchair standards to ensure product quality
for consumers.
2. To participate in the development and revision
of wheelchair standards to provide sufficient
information for product comparison.
Methods Summary
Three basic methods were employed. The first
methods consisted of active participation in the
standards meetings at both the ANSI/RESNA and
ISO levels. This included chairing several of the
working groups, and for two years chairing the
RESNA Technical Guidelines Committee. The second
method was to provide supporting research and
development for the creation and revision of
standards. Without supporting data or devices,
reasonable standards can not be developed. The third
method employed applying the standards to
commercial products to provide comparison data.
This information was published to assist clinicians,
consumers, payers, and manufacturers.
Outcomes Summary
The key outcomes from this task can be
summarized as follows:
• coordinated the development of a complete
electric powered wheelchair/scooter
electromagnetic compatibility standard which is
in the voting process as of 12/98,
• contributed to the development of an electronic
integration interface standard (ISO CD7176/17)
being developed through TIDE, a program of the
European Economic Community,
• conducted a study to compare the results of
common hospital type wheelchairs with active
duty ultralight wheelchairs,
• conducted a study to analyze the performance of
selected lightweight wheelchairs.
• standards which consumers, practitioners,
manufacturers and purchasers can rely upon
more complete information by which to compare
products,
• quality of wheelchairs will be improved.
Recommended Future Developments
Work on the development of standards must
continue in order to ensure improvement in
wheelchairs. Moreover, product comparisons are
required to provide consumers, clinicians, and
manufacturers information about the safety, quality,
and value of wheelchairs. There are a substantial
number of wheelchair standards in development and
in revision. The application of wheelchair standards
continues to produce higher quality wheelchairs.
Publications
Cooper RA, Boninger ML, Rentschler A, Evaluation of
Selected Ultralight Manual Wheelchairs Using ANSI/
RESNA Standards, Archives of Physical Medicine and
Rehabilitation, Vol. 80, 1999.
39FINAL REPORT: 1993-1998
Cooper RA, O’Connor TJ, Gonzalez JP, Boninger ML, and
Rentschler A, Augmentation of the 100 kg ISO Wheelchair
Test Dummy to Accommodate Higher Mass, Journal of
Rehabilitation Research and Development, Vol. 36, No. 1, 1999.
Cooper RA, Gonzalez J, Lawrence B, Rentschler A,
Boninger ML, and VanSickle DP, Performance of Selected
Lightweight Wheelchairs on ANSI/RESNA Tests, Archives
of Physical Medicine and Rehabilitation, Vol. 78, No. 10, pp.
1138-1144, 1997.
Cooper RA, A Perspective on the Ultralight Wheelchair
Revolution, Technology and Disability, Vol. 5, pp. 383-392,
1996.
Cooper RA, Robertson RN, Lawrence B, Heil T, Albright
SJ, VanSickle DP and Gonzalez J, Life-Cycle Analysis of
Depot versus Rehabilitation Manual Wheelchairs, Journal
of Rehabilitation Research and Development, Vol. 33, No. 1,
pp. 45-55, 1996.
Cooper RA, Harmonization of Assistive Technology
Standards , Proceedings 20th Annual IEEE/EMBS
International Conference, Hong Kong, CD-ROM, 1998.
Gonzalez J, Cooper RA, Rentschler A and Lawrence B,
Frame Failures of Welded Tube Manual Wheelchairs,
Proceedings 20th Annual RESNA Conference, Pittsburgh,
Pennsylvania, pp. 184-186, 1997
Lawrence B, Cooper RA, VanSickle DP and Gonzalez J,
An Improved Method for Measuring Power Wheelchair
Velocity and Acceleration Using a Trailing Wheel,
Proceedings 20th Annual RESNA Conference, Pittsburgh,
Pennsylvania, pp. 251-253, 1997
Cooper RA, Gonzalez J, Robertson RN, and Boninger MD,
New Developments in Wheelchair Standards, Proceedings
18th Annual IEEE/EMBS International Conference ,
Amsterdam, Netherlands, CD-ROM, 1996.
Cooper RA, Robertson RN, Boninger ML, A Biomechanical
Model of Stand-Up Wheelchairs, Proceedings 17th Annual
IEEE/EMBS International Conference, Montreal, Canada,
CD-ROM, 1995.
Cooper RA and McGee H, Wheelchair Related Accidents
and Malfunctions, Proceedings 18th Annual RESNA
Conference, Vancouver, BC, pp. 334-336, 1995
Cooper RA, McGee H, Apreleva M, Albirght SJ, VanSickle
DP, Wong E and Boninger ML, Static Stability Testing of
Stand-Up Wheelchairs, Proceedings 18th Annual RESNA
Conference, Vancouver, BC, pp. 349-351, 1995.
top related