Atlas - Examining the Wider Context of Assistive Robotics

H. D. Doran

, S. Neumann

, H. K. Becker

, A. Kollmar

, S. Misoch

, C. Pauli

S. Müller

and F. Hannich

Institute of Embedded Systems, Zürich University of Applied Sciences, Winterthur, Switzerland

Institute of Occupational Therapy, Zurich University of Applied Sciences, Winterthur, Switzerland

Interdisciplinary Competence Centre for Ageing IKOA-FHS, University of Applied Sciences St. Gallen,

St. Gallen, Switzerland

Institute of Marketing Management, Zurich University of Applied Sciences, Winterthur, Switzerland

Keywords: Elderly Care Robotics, Assisted Living, Assistive Robotics.

Abstract: We examine the proposition of a stationary assistive robot arm in the kitchen. Based on a preliminary

business plan and with the aim of generating engineering requirements, a multi-disciplinary project was

established to examine the wider ramifications of such assistive technology in the household, in a Swiss

context, in the fields of health and social wellbeing. Additionally the engineering aspects as well as the

business aspects were examined. We detail both the individual methodologies used in this study, the results

achieved and discuss the results in a wider context.

1 INTRODUCTION

1.1 Motivation

Over the course of a masters level business plan

creation week a group of engineering students

produced a well-argued business plan for a kitchen

assistant robot they called ATLAS. Given the current

interest in assistive robotics but cognisant of the fact

that complete understanding of the application area

was out of scope for the normal robotics engineer it

was decided to construct a multi-disciplinary

consortium to examine the proposal in a more

structured manner. This paper describes the results

achieved (Neumann 2016).

The novelty in this paper is that we focus on the

elderly still resident in their own homes in the

context of enabling them to remain there longer

rather than servicing those in care-homes or those

with constrained movement abilities.

1.2 Consortium and Task

The driving task was to establish engineering

requirements for a robotic assistive arm in the

kitchen guided by the scenario of an elderly person

cooking. The task of the consortium was to establish

whether and in what form a robotic assistive arm was

an acceptable accessory for elderly people in their

household kitchens. The consortium was designed to

complement engineering in domains where

engineering researchers traditionally have no specific

knowledge in this case business, health and social

research. In addition to the research consortium,

expertise from the Swiss national organisation

representing interests of the elderly, Pro Senectute,

was requested and granted. Funding was granted by

the Walder Stiftung.

1.3 Value of Paper and Structure

A large number of assistive robotic projects are

generated by engineering research and less by

market requirements, an exception to this are

assistive robotics for special needs patients. To a

large part, the value of this paper is an indication of

how projects with collaborators from different

disciplines can help generate engineering

specifications or, as in our case, show that an

imagined product might not find a market. We also

show that user requirements are a qualitative and not

quantitative research area and different approaches,

whilst converging on a common solution, offer

greater understanding of the problem domain as well

as otherwise easily missed nuances so this paper also

396

Doran, H., Neumann, S., Becker, H., Kollmar, A., Misoch, S., Pauli, C., Müller, S. and Hannich, F.

Atlas - Examining the Wider Context of Assistive Robotics.

DOI: 10.5220/0006885203960402

In Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2018) - Volume 2, pages 396-402

ISBN: 978-989-758-321-6

serves to illustrate the techniques used by other

disciplines to interested researchers.

This position informs the general structure of the

paper. The next section deals with the engineering

element of this project, Section III, which is further

sub-divided by competence domain, examines the

problem space by that domain. The final section

draws appropriate conclusions and details future

work suggestions.

1.4 Initial Engineering Position

The engineering work consisted of three main tasks,

the establishment of the proposed system capabili-

ties, an advisory position within the consortium and

the development of a mock-up system.

1.4.1 System Proposal

In order to give the other researchers a visualisation

of the system preliminary usability studies were

commissioned. The results, which were presented to

the other researchers, are shown in the figures below.

The initial proposition was that there would be a unit

in the kitchen dedicated to the heavy elements used

in cooking, plates, pounds of sugar or rice etc. and

the oven itself, or more specifically the contents

thereof (Figure 1).

Figure 1: Kitchen setup with assistive robot arm attached

to wall. The vertical shaded area is a kitchen unit designed

for access by the assistive arm. The arm is shown reaching

for a roast chicken in the oven. The shaded horizontal area

is the workspace accessible by the assistive arm.

The morphology of the assistive arm was also

considered important. The initial assessment

considered factors such as number of joints and

structure (Figure 2), the rendering of the arm and

gripper, the form and technology of the gripper

(Figure 3) and whether the rendering would be part

of the control system of the arm or not. Significant

engineering uncertainties were associated with the

rendering of the arm and the gripper, various ideas

from artificial skin to wood were considered before

it was decided to hand this issue over to researchers

from another domain.

2 INTERDISCIPLINARY

RESEARCH

2.1 Further Engineering Work

Independently of the main body of work an

engineering mock-up of a possible assistive arm was

commissioned (Bucher 2016). The work was

temporally staggered with that of the other research

groups. No results of this body of work were shared

with the non-engineering researchers for several

reasons. The first was because of uncertainties of the

design work. Engineering was cognisant of the fact

that because the initial design was starting from the

same knowledge base as the rest of the project,

Figure 2: Possible arm morphologies.

Figure 3: Possible gripper morphologies.

Atlas - Examining the Wider Context of Assistive Robotics

397

several design iterations would be necessary to reach

acceptable price and optic points.

It was also feared that early insight to the

engineering ideas would create a prejudice on the part

of the other researchers who may have less realistic

expectations of the ability of engineering to create an

optically acceptable prototype. Another reason was to

avoid anticipating a solution with the other research-

ers. A fourth was to avoid any specific discussions in

focus groups about the optics of the device. This

mock-up was advanced as far as a simulation and the

drawings and specifications for production of a

prototype including pricing (Figure 4.)

The general features of the assistive arm are a

two-axis arm with a spindle drive for the vertical

direction. The arm can carry 10 kg load and deposit

trays on the work surface on the left of the sink. A

simple hook-like gripper was chosen (Figure 5). The

unit mechanics were designed to allow a self-stow

along the top of the adjoining kitchen elements.

Initial costing for the mechanics, without the

cladding to suit the kitchen design, without the

shelf/oven elements, and without the electronic

control units but including the drives, was estimated

to be around 4’100 CHF calculated with one-off

pricing. Based on these criteria the finished design,

again without cladding and shelf and oven elements,

could retail for about 15’000 CHF.

Figure 4: Simulation of an assistive kitchen arm (far left).

The arm moves up and down and can place/extract trays

from the shelving beside it onto the work surface as well

as move objects in and out of the oven.

2.2 Social Significance of the Kitchen

2.2.1 Motivation

The introduction of a significant technical invest-

ment into the living space may cause a change as the

family learn to live with, and around, such a device.

Electric light or the television are most obvious

examples but the introduction of an assistive robot

arm in the kitchen may, or may not, also represent a

dislocation in the accepted fabric of rooms and their

use by a family and this was the question being

asked in this part of the study. This is a

fundamentally different question as to the acceptance

of such a unit as a technical object as asked in the

next part of the study where the question was based

in reference to the failing physical abilities of the

elderly and the micro-conditions under which such a

unit might be used.

Figure 5: Chosen gripper for the mock-up. There is a drive

integrated in the joint and the end is designed to hook into

the trays and oven door.

2.2.2 Methodology

Two group interviews were organised to establish

the significance of the kitchen in a social context for

the family. The group interview is a structured

interview process which consists of the answering of

pre-prepared open or partially standardised questions

by the interviewees. There is the possibility for the

interviewees to react to and discuss the answers of

the other participants (Misoch 2015).

The first group consisted of two women and one

man between the ages of 70 and 80 all of whom were

married and all of whom have children and

grandchildren. In both cases, the women were

responsible, as homemakers, for kitchen and

household.

The second group consisted of five women and

two men between the ages of 75 and 89. These

participants all had reduced motor-function due to

arthritis, rheumatism, after-effects of strokes, failing

strength in back and arms etc. and were all living in

separate flats in the same housing scheme with each

flat designed for people of that general health

category.

2.2.3 Scope and Results

The three themes researched were the kitchen,

cooking and problems cooking or in the kitchen.

ICINCO 2018 - 15th International Conference on Informatics in Control, Automation and Robotics

398

Kitchen. In all cases the kitchen was considered

more of a functional room than a social room and

like most functional rooms it was important that

everything could be properly stored, and was ready

to be used when required. It represented a

personalised optimisation between aesthetics and

functionality in that kitchen machines that were used

on a daily basis might occupy permanent space on a

kitchen counter. New technology in the kitchen was

in all cases acceptable.

The idea of the purely functional room was often

qualified depending on the available space where

children and husband could talk about the day whilst

the mother was cooking.

The study showed that the meaning of the kitchen

as a social/functional space did not change as much

as the act of cooking during the aging time-span and

that the participants increased their reliance on

kitchen machinery, for instance automated bread-

slicers, as their physical robustness deteriorated as

they aged.

Problems Working in the Kitchen. Issues with

working in the kitchen with increasing age and the

effects, after effects and slow recovery from diseases

could be divided into four categories. Pain whilst

cooking, decreasing strength in hands, arms and

back, changing body sizes and declining strength

during cleaning after cooking.

Pain was often ignored especially when preparing

meals for the partner whereby the scope of the

individual meals and the range of foodstuffs

prepared was significantly reduced, based in part on

the avoidance of pain. The decline in strength, be it

from increasing age or as a result of illness, resulted

in the increased use of electrical and mechanical

appliances and the avoidance of certain heavy

objects, particular pans or pots etc. In both cases the

work-flow, so to speak, was optimised and carried

out at a slower pace to cope with the hindrances.

The changes in body size with age generally

effected the ability to reach things in higher

cupboards and resulted in a rearrangement of the

contents of the storage in the kitchen, which also

resulted, in one case, with the dishwasher being used

to store things.

Finally cleaning after cooking was considered so

problematic that this becomes a reason not to cook

for guests, an effect more pronounced with

individuals who place importance on cleanness and

orderliness.

Cooking. Also questioned was the change of the

importance of cooking as such in a household. In

summary whilst women of that generation may or

may not have experienced fulfilment in cooking, that

function really began with the founding of a family.

The importance of cooking as a social act declines to

a certain extent, especially when the children leave

home, so whilst regular preparation of meals, and

shopping for ingredients, helps structure the day, the

number of complex meals prepared at home tends to

reduce to special occasions. While this also results in

a decline in the ability to manage the cooking of

larger (multi-course) meals it is not seen as a loss

mainly due to the change in appetite and the increase

in available time that reduction of cooking effort

affords. Of course the option of a restaurant for

special occasions was frequently mentioned.

2.3 Acceptance Study

2.3.1 Motivation

Having established that a robotic assistive arm is not

necessarily an intrusion in the personal sphere of the

inhabitants of the dwelling where it is to be installed,

it is important that the articulation of the robot

addresses the needs of the users. Given the

physiological nature of the intended function, in

other words the substitution of specific human

movements due to weakness and pain, health

researchers examined the requirements for

acceptance of a robotic arm in a kitchen

environment.

2.3.2 Methodology

In a first of two work packages, a literature study

was undertaken to identify the difficulties the

elderly, or disabled have while working in the

kitchen and from which requirements on an assistive

robot arm can be established. Then the acceptance

criteria of new technology in the kitchen was

examined and finally a review and critical analysis of

research projects in this field compiled.

The conclusions of this work package were to be

used to inform and guide the constitution and

questioning of focus groups that were to be

organised in the second work package.

The methodology of the first part was to use

databases to find papers on kitchen robotics,

unsurprisingly few papers could be found on this

theme so the search was widened to include general

household robots. After filtering out double hits and

Atlas - Examining the Wider Context of Assistive Robotics

399

those (13) dealing with robot technicalities without

discussing user-experience, 27 papers were included

for further study. From these information and data was

gleaned concerning the problems and limitations of the

elderly in everyday situations, this was used to

generate initial requirements for household/kitchen

robotics. In the third step the acceptance criteria for

technology in household/kitchen robotics was

determined, with particular focus on apprehensions

and fears/suspicions of the target group with respect to

the Technology-Acceptance Model (TAM, Venkatesh

2000).

The methodology applied in the second work

package was to (attempt to) form two focus groups,

one consisting of elderly persons and the other of

people with constrained movement abilities.

Recruitment was very difficult, interestingly enough

those with constrained movement abilities

approached were not interested in technical solutions

believing that they were so constrained that it would

be better for them to learn to live without technical

help. For the elderly focus group only four people

could be recruited.

2.3.3 Scope and Results

Literature Review. The literature review resulted in

the understanding that cleaning help was the most

required assistance in the household (Pigini 2012,

Beer 2012, Ng 2012) followed by reaching for things

and carrying them (Choi 2009, Telson 2013, Beer

2012, Pigini 2012, Fischinger 2016) or holding

things in an assistive capacity (Pigini 2012, Beer

2012). Since potential users, through the literature,

generally accept that that a robot could be useful in a

household, the activities with the planned focus

groups would serve to confirm the literature and help

inform the necessary acceptance criteria which can

be structured according to the TAM.

The TAM informed criteria for questioning the

focus groups – in this case perceived usefulness;

perceived ease; image and technology experience.

The TAM does not recognise differences between

the sexes but Brodbent et. al. (Brodbent 2009) find

substantial difference to what they attribute to the

more emotional approach of women vis a vis the

more clinical approach by men towards technology.

Focus Group(s). Severe difficulties were encounter-

ed in trying to form two focus groups and eventually

only one of seven members could be formed. The

results of the literature could only be partially

predicted. Far more value was placed by the

participants on the difficulties of opening things

(bottles, jars, tins) than moving things from A to B.

In fact a long list of complaints about current bottle

and tin-openers followed. So in terms of the category

perceived usefulness a proposed kitchen assistive

robot was given a task list that it should be able to

perform which began with opening things, carrying

things through cleaning tasks and ending with

peeling vegetables.

The “image” results were less clear. Image in this

context is the effect on a person’s self-image of

being dependent, in this case, on technology. The

one participant with movement restrictions was more

inclined to want to use a robot for a multitude of

tasks whereas most other participants made it

dependent on limited available third-party human

resources – if human help was available however

most would prefer a human to do it.

In terms of technology experience most

participants had enough contact with technology not

to be put off, under the conditions already

mentioned, to consider using it.

The group also considered the practical issues

including general functional characteristics, costs

and general look and feel. It was clearly seen as a

kitchen device and should be in the same category as

a dish washer and/or cooker, it should be able to put

itself away after performing its task and be stationary

not mobile and it should look technical, the idea of it

having human characteristics wasn’t welcomed.

2.4 Business Considerations

2.4.1 Motivation

The business researchers examined the economic

viability of such a product and attempted to identify

the main influencers, those who, in this case, would

influence a potential user to purchase, in this case, an

assistive robot arm.

2.4.2 Methodology

A marketing methodology, under the assumption of a

strategic decision to market a kitchen assistive arm,

was applied. The marketing methodology consists of

several discrete steps generating measureable output

and culminating in the well-known marketing mix of

Price, Product, Place and Promotion (four Ps (Doyle

2006)). Apart from generating guidelines for the

marketing and sales of any technical unit, or washing

powder for that matter, the four Ps can be used to

generate engineering specifications right down to

ICINCO 2018 - 15th International Conference on Informatics in Control, Automation and Robotics

400

packaging, logistics and maintenance policies. The

derivation of the four Ps is via defining a market

development strategy, which in turn is derived from

determining the market aims. The marketing aims

define the demographics, market segments, position-

ing of a product with respect to potential competitors

2.4.3 Scope and Results

From the point of view of an assistive robot arm for

kitchen use the market segment are the elderly and

disabled, the former the larger market by far. The

potential market size was determined by the use of

official statistics qualified by consideration of

various factors. These factors are, amongst others,

the aging characteristics of the target segment, in

other words when and how severely health declines

to a point where such an assistive technology

becomes useful and the duration of residence in the

own-flat before moving to a third-party run residence.

In particular the age from 80 seems to be a critical

period (in Switzerland) which sees a rapid increase in

the requirement for specialised care (Höpflinger 2011,

BfS 2015). Not only is that but the rate of degenerati-

on once a health dislocation has occurred is also rapid.

The ramification of this is that the market segment for

an assistive robot is the age group from 65-80 but the

window of opportunity for the installation of an

assistive robot would appear to be small.

A further hindrance is, at least in the canton of

Zürich, that the average 60 year old has lived in the

same accommodation for 30 years typically in 3-4.5

room apartments, which tend to be relatively small in

Switzerland at 100 m

. The length of time lived in

the same space would tend to discourage complica-

ted and inconvenient retro-fits, especially difficult

and expensive in smaller kitchens (Figure 6).

Figure 6: Typical kitchen layout in an older Swiss 3-4

room apartment (comparis.ch).

Whilst the stairs and the bathroom were

considered by the target group to be the main

hindrance in a currently occupied apartment, 36% of

the respondents of a study categorised the kitchen as

also being unsuitable for them (Höpflinger 2014).

The conclusion to be drawn here is whatever the

functionality of an assistive arm, the installation

aspects must be well considered.

The positioning of the solution with respect to

other assistive products, both high- and low-tech and

the strategic pricing options were also analysed.

Unsurprisingly there appear to be no potential

competitors on the market, except possibly for

Moley from the UK, which intends to offer a

complete programmable cooking station (Moley

2015).

Finally the sales channels and, most importantly,

the potential influencers were analysed. In terms of

influencers it was quite clearly determined that

partners and children are the main influencers, in

other words effort must be made to gain acceptance

of (grown-up) children in order to sell such a device.

3 DISCUSSION AND FUTURE

WORK

3.1 Discussion

There are two foci of comparison of the studies

undertaken. The first is the comparison of the results

of the non-engineering studies with each other and

the second the combined results with the output of

the initial engineering mock-up.

Whilst the three non-engineering studies

complement each other they also individually offer

nuances which otherwise might have gotten

unnoticed and so, a more complete picture of a

situation can be built. One theme is the declining

importance of the kitchen as an area in the context of

a family as the offspring leave the house. The kitchen

tasks are reduced in their complexity and frequency.

Despite this reduction in importance the kitchen is still

seen as a necessary workroom by the elderly and

whilst they are, prepared to invest somewhat in

equipment, they exhibit health and strength declines

that makes the use of such a robot merely to move

stuff from A to B an underutilization of a potentially

expensive piece of equipment.

The consequences for engineering is that a more

holistic approach to a kitchen assistive robot than

what was initially conceived as a pick-and-place

Atlas - Examining the Wider Context of Assistive Robotics

401

robot is required. The assistive unit must be able to

help grip things; jars, bottles, tins and the like; assist

in opening them; pour the contents into a pot it has

moved from storage to the stove; clean the tin for

disposal (common in Switzerland); move pots and

pans around the stove and then assist in cleaning the

pots and pans when the cooking activities are

finished. In this light, what is required is a sort of

exo-kitchen machine rather than an assistive help for

the elderly, which might even help dispel the image

issue raised by the acceptance study. The three non-

engineering studies clearly gave sufficient

information to help understand a kitchen situation in

a more structured fashion.

The three non-engineering studies also provided

sufficient guidelines for engineering to design the

look and feel of such a robot. A technical look-and-

feel is acceptable meaning normal kitchen standards

could be applied and, in our opinion, standard

usability features from industrial collaborative

robotics could be applied in adapted form. We also

believe that the mock-up could probably be

considered close to a useful first mechanical

prototype.

3.2 Future Work

As unaware as engineering professionals may be

about the methodologies employed by non-

engineering researchers so are non-engineering

researchers unclear about what engineers need in

order to understand end-user requirements. It might

be worth establishing a methodology to allow non-

engineering researchers to ensure that feedback from

focus groups in their research domain is as

unequivocal as possible.

The research and development of an assistive

robot and its use in real-world situations is not

impossible but its success in the open market would,

based on the presented results, be challenging. Re-

framing the issue scope, as suggested in the previous

sub-section, along the lines of a kitchen machine

might mitigate the negatives out of the presented

study and accentuate the positives for surely, if such

a kitchen machine becomes standard issue the

elderly will benefit as well.

ACKNOWLEDGMENTS

Thanks are due to the Walder Stiftung, Zurich, for

their financial support: to Pro Senectute, Zürich for

their support: and to R. Pfrommer and H.

Hesselbarth, Institute of Mechanical Systems,

ZHAW, for organizing and managing the

engineering mock-up project. Lukas Reisinger,

Institute of Embedded Systems, ZHAW, was

responsible for the initial arm and gripper designs.

REFERENCES

Beer, J. M., Smarr, C. Chen, T. Prakash, A. Mitzner, T.

Kemp, C. Rogers, W.A. 2012 “The Demesticated”.

BfS, Bundesamt für Statistik. “Statistik der Hilfe und Pflege

zu Hause - Ergebnisse 2014: Zahlen und Trends.“ 2015

Neuchâtel: BfS.

Brodbent, E., Stafford R., MacDonald B., 2009 “Acceptance

of Healthcare Robots for the Older Population: Review

and Future Directions,” Int J Soc Robot 1:319-330.

Bucher, S. Di Fazio, L. “Kitchen-Aid”, 2016. Project report

ZHAW, unpublished.

Choi YS., Deyle T., Chen T., Glass J.C., Kemp CC. 2009

“A List of Household Objects for Robotic Retrieval

Prioritized by People with ALS”, IEEE, 11th

International Conference on Rehabilitation Robotics,

Kyoto International Conference Center, Japan, June

23-26.

Doyle, 2006 Marketing Management and Strategy, Prentice

Hall; 4 edition (May 28, 2006)

Fischinger D., Einramhof P., Papoutsakis K., Wohlkinger

W., Meyer P., Panek., Hofmann S., Köertner T., Weiss

a., Argyros A., Vincze M., 2016. Robotics and

Autonomous Systems 75; 60-78.

Höpflinger, F., Bayer-Oglesby, L. & Zumbrunn, A. 2011

Pflegebedürftigkeit und Langzeitpflege im Alter. Bern:

Hans Huber.

Höpflinger, F., Van Wezemael, J. (Hrsg.) 2014. Age Report

III: Wohnen im höheren Lebensalter Grundlagen und

Trends. Zürich: Seismo.

Misoch, S. 2015 “Qualitative Interviews.” Berlin: Walter de

Gruyter.

Moley 2015 http://www.moley.com/ last accessed

16.03.2018

Neumann, S., Becker, H. K., Kollmar, A., Misoch, S., Pauli,

C., Doran, H. D., Müller, S., Hannich, F. 2017.

«Küchenassistenzroboter für Seniorinnen und Senioren:

Bedürfnisse, Akzeptanzfaktoren und Wirtschaftlichkeit.”

In: G. Kempter, I. Hämmerle, (Ed). Umgebungsunter-

stützes Leben. Beiträge zum Usability DayXV (38-44).

15. Lengerich: Pabst.

Ng, J. Tan, O. Wong, A. Kiat, K.W. 2012. “Older Adults'

Attitudes towards Homfes Service Robots,” WASA, 11;

26-27 2012.

Pigini, L. Facal, D. Blasi, L. Andrich, R. 2012. “Service

robots in elderly care at home: Users' needs and

perceptions as a basis for concepts development,”

Technology and Disability 24; 303-311.

Telson J., 2013, Research Update- A Helping Hand in the

Kitchen; PN, 38-40

Venkatesh, V. Davis, F. 2000 “A theoretical extension of the

technology acceptance model: Four longitudinal field

studies.” Management science 46(2), 186–204. 2000.

ICINCO 2018 - 15th International Conference on Informatics in Control, Automation and Robotics

402