RESNA 26th International Annual Confence

Technology & Disability: Research, Design, Practice & Policy

June 19 to June 23, 2003
Atlanta, Georgia

Smart Power Assistance Module for Manual Wheelchairs

Richard Simpson1, 2, 2, Songfeng Guo1, Dan Ding1, 1
1Department of Rehabilitation Science and Technology, University of Pittsburgh, Pittsburgh, PA
2AT Sciences, Pittsburgh, Pa

Abstract

Investigators are developing a system of components that can be attached to manual wheelchairs to convert them into "smart" manual wheelchairs. The system is being designed as a mobility solution for individuals with visual and physical impairments, but is expected to be useful for other user populations as well.

Background

The concept of power assistance for a manual wheelchair is relatively new, and represents a viable alternative for individuals who are unable to generate sufficient propulsion force to use a manual wheelchair, but do not wish to use a traditional powered mobility device [1]. In a power assisted manual wheelchair, the traditional rear wheel hubs are replaced with motorized hubs that serve to magnify or reduce (i.e., brake) the propulsive force applied to the rear wheels by the user. Investigators are using power assistance as the basis for a Smart Power Assistance Module (SPAM) that provides independent mobility to non-ambulatory individuals with visual impairments. The SPAM will use sensors to detect obstacles and drop-offs (e.g., stairs, curbs, potholes) near the wheelchair, and use the motorized hubs to modify the forces applied to each wheel to avoid obstacles.

Statement of the Problem

The target user population for the SPAM consists of individuals with both a visual impairment and a mobility impairment that makes it difficult or impossible to ambulate independently using a white cane, guide dog, or other traditional mobility aid for the visually impaired. The American Federation for the Blind (AFB) has estimated that 9.61% of all individuals who are legally blind also use a wheelchair or scooter, and an additional 5.25% of individuals who have serious difficulties seeing (but are not legally blind) also use a wheelchair or scooter.

Design

The SPAM is based on the Yamaha JWII Power Assist Wheelchair Hubs (shown in Figure 1), which were recently approved by the FDA [1]. The JWII hubs meet the mobility needs of a population of users that (1) need a mobility aid, (2) lack the upper-body strength or function to propel a manual wheelchair, and (3) do not want a powered mobility device.

Figure 1. Yamaha JWII Power Assist Wheelchair Hubs
Figure 1. Yamaha JWII Power Assist Wheelchair Hubs. The left half of this figure shows an overhead view of the components of the JWII system. The JWII hubs measure pushrim forces using linear compression springs and a simple potentiometer that senses the relative motion between the pushrim and the hub. A microprocessor uses these signals to control a permanent magnet direct-current motor attached to each rear wheel. Each motor is connected to a ring gear affixed to a hub of each wheel with spokes radiating out to the wheel rim. Power is supplied by either a single custom designed nickel-cadmium battery (NiCd) or a nickel-metal hydride battery (NiMH). The wheels use ball-lock pins (12 mm in diameter) for easy removal. A secondary pin, offset from the axle, is used to provide the reaction torque at the frame in response to the motor. The right half of this figure shows a wheelchair with the JWII system attached.
filler

 

Figure 2. Schematic for unmodified JWII system (left) and SPAM (right)
Figure 2. Schematic for unmodified JWII system (left) and SPAM (right). The left half of this figure shows the flow of information within the JWII system. The force applied to the JWII hubs is translated by sets of linear compression springs and recorded by a simple potentiometer that senses the relative motion between the pushrim and hub. The microprocessor receives the potentiometer signals and coordinates control of each wheel's permanent magnet direct-current motor. The right half of this figure shows the flow of information within the SPAM. In the SPAM, the force applied to the hubs is received by an additional computer, which combines this information with feedback from the SPAM's sensors to modify the wheelchair's direction of travel. This modified signal is then passed on to the JWII system's microprocessor.

The right side of Figure 2 shows the design of the SPAM. The SPAM will be able to sense (1) the propulsive force applied to each rear wheel of the wheelchair, (2) the magnitude and velocity of rotation of each rear wheel, and (3) the location of obstacles and drop-offs relative to the wheelchair. Information from all sensors will be collected by a computer which will: (1) run the system's control software, (2) integrate information about the user's input and the surrounding environment, and (3) pass command signals to the JWII system's microprocessor.

Several types of sensors will be integrated into the SPAM prototype. These sensors will be used for (1) tracking the state of the wheelchair (e.g., wheel velocity, torque applied to each rear wheel by the user) and (2) locating obstacles and drop-offs in the wheelchair's environment. The JWII hubs already contain sensors for measuring the propulsion force applied by the user to each wheel (potentiometers) along with sensors for measuring rotation of the rear wheels (Hall Effect Sensors).

Obstacles are identified using sonar and infrared range finders. The sonar sensors (Sonaswitch Mini-A, EDP Company, Livonia, MI) have a maximum range of 2.4 m (7.9 ft) and a minimum range of 21.8 cm(8.6 in). The advantages of a smaller range are that (1) the frequency of sonar readings is increased and (2) the sonar system will be able to see obstacles that are extremely close to the wheelchair, which is important for passing through doorways. Infrared range finders provide a focused, highly modulated infrared beam, providing absolute ranging based on simple triangulation. The result is an accurate range value between 1.7 and 0.1 meters in a variety of circumstances for the Sharp GP2Y0A02YK (Sharp Electronics, Mahwah, NJ). Bump sensors serve as the ``sensors of last resort'' for the SPAM. When a bump sensor is activated it will bring the chair to an immediate halt. Bump sensing is implemented using contact switches placed on the leading edges of the wheelchair.

Table 1. Control Parameters

Parameter

Range

Description

Maximum Gain0\% - 67\% Amount of gain that SPAM applies to force applied to pushrims by user.

 

 

Maximum Speed

0 m/s - 1.2 m/s

Maximum speed achievable by SPAM. 1.2 m/s is average walking speed.

Minimum Obstacle Clearance2 in to 12 in Minimum distance between obstacles and wheelchair maintained by SPAM.

 

 

The SPAM control software will provide smooth travel while modifying the direction and speed of travel of the wheelchair to avoid obstacles. The control software will combine the existing JWII control software with obstacle avoidance software designed for smart power wheelchairs [2]. Changes in the wheelchair's direction will be accomplished by causing one rear wheel to spin slower than the other, either by accelerating the opposite wheel or through the use of the JWII's regenerative braking capability. Table 1 lists some of the parameters of the control software that will be used to determine the behavior of the SPAM.

Significant attention will be paid to developing a user override mechanism that allows the user to quickly and easily disengage the SPAM's obstacle avoidance in an emergency, but is not so ``quick and easy'' that the override mechanism can be engaged inadvertently. Furthermore, the behavior of the system (both the SPAM components and the Yamaha JWII hubs) once the SPAM is disengaged will be explored to identify how to maintain the user's safety.

Development

At present, the investigative team has identified the sonar and infrared sensors that will be used on the SPAM for obstacle detection, and is conducting experiments to characterize the behavior of the JWII system. Various approaches are being considered for providing navigation assistance to the user. Additional development will include examining solutions for drop-off detection, identifying an appropriate microprocessor, and developing mounting solutions for attaching the components of the SPAM to a range of manual wheelchair frames. Finally, we plan to conduct testing under controlled laboratory settings and out in the field.

Discussion

The SPAM is intended to be an ``after-market addition'' to Yamaha's JWII hubs, which will take advantage of recent efforts which changed the connection between the hubs and the processor unit from a hardwired connection to a plug-and-jack connection. In addition to serving the needs of the visually-impaired, the SPAM is also expected to be useful as an intelligent one-arm drive mechanism for hemiplegics.

References

  1. R. Cooper, S. Fitzgerald, M. Boninger, K. Prins, A. Rentschler, J. Arva and T. O'Connor (2001). Evaluation of a Pushrim Activated Power Assisted Wheelchair. Archives of Physical Medicine and Rehabilitation, 82(5): 702-708.
  2. R. Simpson, D. Poirot and F. Baxter (2002). The Hephaestus Smart Wheelchair System. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 10(2): 118-122.

Acknowledgments

This research is funded by a Phase I Small Business Innovation Research grant from the National Eye Institute (#1R43EY014490-01).

Richard Simpson
Department of Rehabilitation Science and Technology
University of Pittsburgh; Forbes Tower,
Suite 5044; Pittsburgh, PA 15260
ris20@pitt.edu

RESNA Conference Logo