The Development of RAIL
(Robotic Aid to Independent Living)
Mike Topping (1)
Helmut Heck (2)
Gunnar Bolmsjo (3)
David Weightman(4)
Staffordshire University
School of Art & Design, Centre for Rehabilitation Robotics,
Stoke-on-Trent, Staffordshire, UK.
tel: +44 1782 294477
fax: +44 1782 294414,
email: Rehabrobotics@Compuserve.com
1. Summary
The aim of the RAIL project is to further develop the Handy 1 rehabilitation robot [1] to enable severely disabled people to gain/regain independence in several important daily living activities such as: eating, drinking, washing, shaving, teeth cleaning and applying make-up. To-date no single system, robotic or otherwise exists that is capable of assisting severely disabled people to gain freedom in these tasks.
Over the past 20 years, there have been many attempts at developing robotic devices to improve the lives of people with disabilities. The most significant commercial developments and achievements in rehabilitation robotics to date, can be summarised as follows:
In the USA, Larry Leifer Stanford University designed a robotic workstation, De VAR (Desktop Vocational Assistant Robot), which, under voice control, enabled computer programmers with disabilities to return to work. The cost of this system at around £50,000 is expensive.
In Europe, the main commercial developments have come from Kwee (Netherlands), who designed a demountable wheelchair robotic manipulator system- the MANUS teletheses. This is a purpose built robotic arm, used by disabled people to pick up and manipulate objects. The cost of this system is around £30,000.
The 'Handy 1' developed in the UK by Mike Topping is the world's most successful and widely used rehabilitation robotic system to-date [2], with over 120 units placed. Originally designed to provide independence at mealtimes to the severely disabled, its cost, at £4,000, is low in comparison to all other commercial developments.
| Top|
2. Introduction
In order to achieve the improved functionality needed for the RAIL project it was found that the existing Handy 1 system required further development and upgrading in the following areas:
2.1 Design of new control system
The specification for a new control was completed which comprised of three main components a processor board, a main input/output (i/o) board and a daughter board.
The main i/o board was designed to interface to the PC/104 range of target systems. The processor board chosen was the 80186 based TB8680 model. The standard DOS found on IBM compatible PC's can be loaded and run from EPROM on the PC/104 systems thus simplifying program development, also widely available development tools can be utilised.
A simplified 8 bit design was chosen to ensure that most elements of the specification could be incorporated on the main i/o board. Surface mounting was ruled out as too expensive, and 16 bit tracking would have consumed too much board space.
The connections between the PC/104 plugs and sockets were reduced to the absolute minimum that were required for the daughter board design. The supply used is 12v, which is required to drive the stepper motors and the electromagnet, the 12v is then regulated down to 5v for the processor and the i/o boards. Briefly the elements of each board are described as follows:
- Main i/o board
- A serial port, 64 LED drivers, 8 outputs for 12v 500mA equipment, 8 stepper motor drivers, 16 TTL inputs, 8 TTL general inputs and 6 programmable timers.
- Daughter board
- Joystick port, 8 bit digital to analogue converter and amplifier and hardware timer. The TB8680, 1 Centronics bi-directional parallel port, 3 RS232 serial communication ports, one 12 bit a/d converter, keyboard interface and speaker interface.
2.2 Design of detachable tray sections
The current eating and drinking tray attachment arrangement needed modification to allow inter-changeable modules for washing, shaving, teeth cleaning, make-up, as well as the existing eating and drinking to be incorporated, thus facilitating a greater variety of tasks to be performed by the disabled user.
Specifications were produced for the attachments, these take the form of three detachable slide-on trays. Each tray is of an aesthetically pleasing design which is ergonomic, hygienic and robust in construction. Tray 1 combines the eating and drinking function, tray 2 the teeth cleaning, shaving and washing functions, a hot air face dryer is incorporated in the design, tray 3 facilitates the make-up option, allowing ladies to apply a range of cosmetics including: blusher, lip stick, eye shadow and foundation. A simple LED control module combined with a single switch input is incorporated in all three tray designs.
|
|
|
|
|
fig. 1 The eating/drinking tray |
fig. 2 The washing/teeth cleaning/shaving tray |
|
|
|
||
|
fig. 3 The make-up tray |
2.3 Writing of C++ program code
The existing Handy 1 control system is based around a Z80 assembler which for the purpose of the RAIL project was thought impractical and replaced with PC104 technology, therefore complete code for all Handy 1 attachments needed to be re-written in C++ code. The detachable trays for the RAIL system have been designed to rest upon runners attached to the Handy 1 and connect electrically via a robust sixteen pin socket built integrally into the Handy 1 base unit. Using this principle it is possible for the system to recognise the presence or absence of up to fifteen different trays. Feedback from the joints of the Handy 1 is incorporated to enable automatic referencing on start-up, also a simple error checking routine is built in to the solution. Ten turn potentiometers have been mounted via gearing to each joint of the Handy 1 to facilitate the feedback and C++ code has been generated.
2.4 User requirements analysis
In order to achieve the above technical developments close collaboration with potential users was maintained throughout the project. The user requirements analysis commenced with the identification of a user group, also a detailed study was carried out with the existing Handy 1 user group so as to:
- gain experience with the current Handy 1, in order to make suggestions for improving if necessary the existing eating and drinking functions.
- analyse in detail the user requirements regarding the proposed washing, shaving, teeth cleaning and make-up applications.
- peer review the documents collected during the user requirements study.
Members of the user group included:
- current Handy 1 users and their care workers.
- primary users being disabled people working in the sheltered workshop at Evangelische Stiftung Volmarstein (ESV), disabled students, a disabled assistant professor at the University of Cologne.
- people testing the system for its ease of use and user friendliness included children aged between 8 and 10 years old.
- secondary users included rehabilitation personnel within the sheltered workshops at ESV, rehabilitation engineers at FTB and staff at the University of Cologne.
The user studies at ESV were conducted during lunch breaks which gave an opportunity to evaluate the Handy 1 with a large number of different menus eaten in a typical working eating environment for people with disabilities. In contrast to this the users at the University of Cologne used the Handy 1 in their private home environments where a large variety of different foods were tested.
The tests yielded suggestions for improvements and set the basis for the discussion and analysis of the requirements for the new applications.
| Top |
3. Principle of Operation
The R.A.I.L system has the capability to assist severely disabled in (re)gaining independence with several everyday tasks including: eating, drinking, teeth cleaning and shaving. The disability groups that could benefit from this development include: cerebral palsy, motor neurone disease, muscular dystrophy, multiple sclerosis, stroke and degenerative illness of the elderly [3].
The RAIL functions (eating, drinking, shaving, teeth cleaning, washing and applying make-up) are designed to be attractive, retrofittable modules capable of being used individually or in combination. The capability to achieve this is important if the system is to have sufficient flexibility to be used by people with a range of different disabilities. The attachments may therefore be simply added or removed to accommodate either gains or losses in an individual's physical capabilities. This modular design criteria will also allow for the future development of additional attachments [4].
|
|
|
|
|
Eating and Drinking |
Washing/teeth cleaning/shaving |
Make-up |
fig 4. The RAIL system with its various tray attachments
3.1 Human-Machine Interface
The R.A.I.L system's human-machine interface takes into account the differing input and control requirements of disability groups mentioned earlier. The interface specification includes single switch input options and voice recognition capabilities for future development.
3.2 Control of the R.A.I.L
The RAIL control module utilises a LED scanning control, this method is now widely used by people with special needs to control their environment. LCD control panel capabilities are also designed into the systems specification for future development.
3.3 Height Adjustment of the R.A.I.L
The RAIL system is fully height adjustable to allow the robot to be used by both children and adults, all sitting at different heights. This is achieved by using a simple gas spring method, enabling the system to be easily adjusted through a significant range of movement to cater for different seating positions and heights.
3.4 End-Effector
The RAIL system is fitted with a single end-effector design which is capable of manipulating all the attachments. The solution is low cost, effective and based around peg-and-cup methodology. The peg is designed to incorporate a integral 12v electro holding magnet, thus affording the robot the capability of rotating the attachments, where necessary, through 360 degrees, without dropping them. The cup section has an armature plate fitted internally and each cup section is designed to be an integral part of the design of each attachment.
| Top |
4. Overview of the potential for the RAIL system
Changing age structures, resulting in increased numbers of people with special needs are making ever greater demands on the community of care workers. Dependency upon care staff, particularly in public institutions, where volume dictates the level of personal attention, can have a significant effect on the well being and quality of life of the individual. The introduction of systems such as RAIL will encourage greater personal activity, leading to an increased level of independence.
The impact of the R.A.I.L project on the community of care workers will also be significant helping to reduce the amount of stress present in situations where care workers assist disabled people on a one-to-one basis [3].
The R.A.I.L system will be particularly beneficial in rehabilitation centres and institutions for the elderly, where staff have to care for many people. R.A.I.L will, for example, enable people to eat simultaneously, as opposed to being fed one after another by care workers. Furthermore, personal hygiene will become possible with a lower level of assistance [3].
The development of the R.A.I.L project bears world wide relevance, adding strength to the European scientific basis for competitiveness in the area of rehabilitation robotics. The consortium already has a great deal of experience regarding user requirements and expectations for rehabilitation robotics and the RAIL programme has enabled us to further extend this knowledge, to the ultimate benefit of the European disabled community.
Although pre-competitive in nature, the project is ultimately market-orientated; the consortium intends to commercially exploit the results of the R.A.I.L project by producing a high quality working system, capable of being a market leader, not only in Europe, but also in Japan and the USA.
| Top |
5. Conclusions
Rehab Robotics Ltd. are already successfully pioneering the marketing of rehabilitation robotics, with their 'Handy 1' robot and significantly the market for this system is slowly emerging.
It has been shown by Johnson and Deitz [5] that on average, around 3.5 hours per day are spent assisting people with special needs. This relates specifically to eating and does not take into account the hours spent attending to matters of personal hygiene, such as teeth cleaning, shaving, grooming, etc.
The potential for a R.A.I.L system costing around £6000, will be significant for a number of reasons including:
- Unique, with no competitors
- Low unit cost
- High return on investment from time/cost released to other activities
- Improved quality of care provision
It is now planned that the introduction to the world market of the RAIL system will follow on from the success and penetration of the 'Handy 1' system and will pursue similar marketing methodologies at an increased level of activity.
| Top |
6. Acknowledgments
The RAIL project is grateful for funding and support from the European Commission's DGXII BIOMED 2 Programme.
| Top |
7. Partnership
Staffordshire University, UK - Coordinator. (4), FTB, Germany (2), Lund University, Sweden (3), Rehab Robotics Ltd., UK (1).
| Top |
8. References
[1] Topping M J (1996) Handy 1, a robotic aid to independence for severely disabled people. Published in Technology and Disability 5 (1996) pp233-234
[2] Weir, RFff, Childress, D.S. (1996) Encyclopaedia of Applied Physics, Vol. 15
[3] Topping M J, Helmut H, Bolmsjo G, (1997) An overview of the BIOMED 2 RAIL (Robotic Aid to Independent Living) project. ICORR'97 International Conference on Rehabilitation Robotics, 14-15 April 1997, Page 23 - 26.
[4] H. Heck, Ch. Buhler, P. Hedenborn, G. Bolmsjo. M. Topping, (1997) "User requirements analysis and technical development of a robotic aid to independent living (RAIL). 4th European Conference on Engineering and Medicine Bridging Eat and West - Warsaw (Poland) 25-27 May 1997.
[5] Johnson and Deitz , American Journal of Occupational Therapy Vol.39, 578-583
| Top | | TIDE 98 Papers |