SERVICE ROBOTS FOR THE ELDERLY ?

 

Carl-G嗷an Hed始, chairman organizing committee:

SUMMARY REPORT ON BIOPOLICY SEMINARS 2004/2005.

 

 

                                                                                   cf. chapter 12:3

 

 

1.Venues.............................................................................................................................. 3

2.Organization................................................................................................................. 3

3.Support............................................................................................................................ 3

4.An experiment in transdisciplinary networking............................................. 4

5.Preparations for the 2005 Seminar......................................................................... 5

6.The seminarsﾕ roots in biotechnology............................................................... 7

7.Comments from the Internet discussion........................................................... 11

8.The Voice of experience, interview: Joseph Engelberger.......................... 18

9.The Voice of knowledge: videodialogue Stockholm -Tokyo.................... 28

10.Local experiences and the European dimension........................................... 33

(afternoon session at IVA)......................................................................................... 33

11.The Biopolicy Award............................................................................................... 36

12.POST SCRIPT on a possible follow-up............................................................... 40

REFERENCES...................................................................................................................... 47

Figures.............................................................................................................................. 50

ENCLOSURES...................................................................................................................... 53

 

 

 

1.Venues.

The 2004 seminar took place at the Royal Swedish Academy of Sciences (KVA) on October 27, and the last conference at the Royal Swedish Academy of Engineering Sciences (IVA), Sept.21/22 2005. The latter constituted the final endeavour of the Biofocus Foundationﾕs (BFﾕs) ﾓBiopolicy Seminarﾓ -series, since the Foundationﾕs Stockholm Office closed on October 1st (BF is now headquartered in Dublin, with a regional office in Dubai).

 

 

2.Organization.

The seminars were organized under the auspices of the World Academy of Art and Science (WAAS) and in cooperation with the Engineering Sciences Division (VIII) of KVA, and the Biotechnology Division (X) of IVA. Those bodies, as well as the Karolinska Institute (KI) and the Royal Institute of Technology (KTH) were all represented on the organizing committee.

 

The committee had the following composition: Walter Truett Andersson (WAAS, USA), Gustaf Brunius (X), Gunnel Dalhammar (KTH, only 05), Geoffrey Hamer (BF, Ireland), Carl-Goran Hed始 (Committee chairman, VIII), Gosta Lagermalm (X), Tommy Jonsson (Secretary, BF), Lennart Levi (KI, only 05), Sam Nilsson (WAAS), Erling Norrby (KVA), Mathias Uhlen (X), Bengt-Arne Vedin (IVA, only 04) and Anders Wijkman (BF-Chm.,EU Parl.). For adresses to those individuals or others mentioned in this report please contact <cgheden@biofocus.org>.

 

 

3.Support

The Seminars were kindly  supported by the Biofocus Foundation, ABB, DeLaval, Grimaldi Industri AB (05), IKEA, JM, Mellon Holdings, Riksbankens Jubileumsfond, Sven & Dagmar Sal始ﾕs Stiftelse, Vinnova, Wennergrenﾕs Stiftelse (05) and ］gpannef嗷eningens Forskningsstiftelse(05).

 

The preparatory Internet discussions were ably managed in 2004 by Mr.Eng Leong Foo and Dr Burke Zimmerman (WAAS,USA), and in 2005 by Mr.Tommy Jonsson (BF) and prof. Lennart Levi (KI).

 

Planning support by prof. Henrik Christensson (KTH) for the swedish component of the video-dialogue on Sept.22nd 2005 is gratefully acknowledged, as is also coordination of the Tokyo inputs to this dialogue by professor Takao Nuki of the Science Council of Japan and professor em. Hirochika Inoue, Inspector General of the Japan Society for Promotion of Science. In Tokyo communication facilities and IT-staff was kindly provided by professor Kazuhiko Nishi.

 

 

4.An experiment in transdisciplinary networking.

 

Like earlier seminars, the 2004 and 2005 conferences had a focus on long-term technology assesment- and on transfer-issues related to sustainable development. However, there was a special link between those two seminars: Vinnovaﾕs supporting grant: ﾓSocial Innovations for Development – New Visionsﾓ. This concerned an effort to compare two computer-aided conferencing techniques designed to stimulate bioengineers to think about social innovations in response to major innovations or problems with a global reach. Are there perhaps simple electronic shortcuts that can produce the same intellectual yield as the corridor-comments about the plenary adresses at major international congresses?

 

The 2004 seminar (ﾓSocial Innovations for Development – a Case for Biopolicyﾓ) had its roots in a 1979 seminar with the same thrust (Heden and King,1984). 25 years later this caused the observation that the rate of change had obviously been so great that spontaneous adaptive innovations had not been able to cope with various risks and social stresses. It was felt that, even if the way ahead for anticipatory social innovations would have to follow a trans-disciplinary roadmap, biotechnology might serve as a useful signpost.

 

However, a basic question early emerged: what might be the best way to make experts in biotechnology step outside the borders of their disciplines (genetic engineering, immunology, microbial ecology, process engineering etc.)? How could they be induced to take the time to discuss the global aspects of the ethical, economic and health dimensions of those disciplines?

 

In order to adress this question the 2004 and 2005 seminar programs were based on topics of general interest (cf. encl.A,B) prepared for one-day seminars by means of month-long Internet discussions. In 2004 those discussions concentrated on ﾓLearningﾓ and on various ﾓChallenges to Personal Integrity and Privacyﾓ. They got a flying start from IVA:s contribution to ECﾕs Science Generation Project: an international conference on ﾓBiotechnology, Possibilities, Risks, Ethics and Societyﾓ. However, the invitations were also broadcasted internationally over many networks, such as the International Organization for Biotechnology and Bioengineering /IOBB/ and the UNEP/UNESCOﾕs Microbiological Resource Centers.

 

Also the 2005 seminar (ﾓEngineering to Improve the Life Quality of the Elderlyﾓ) had deep roots, but this time  in the Biofocus Founda-tionﾕs longstanding interest in the impact which demographic and environmental factors might have on sustainable building designs (small schools and nursing homes).

 

However, there were significant differences between the two seminars. The preparatory Internet discussion for the 2005 seminar (www.biofocus.org/phpBB2), for instance focused on a narrower theme: ﾓPros and Cons of Service Robots for the Elderlyﾓ and on a limited number of experts. Also the discussions were designed to get the most out of an online debate between two countries (Japan and Sweden) with ageing populations but with great differences in traditions and culture.

 

 

 

 

5.Preparations for the 2005 Seminar.

 

The preparations for the last seminar started in the spring of 2004, i.e. well ahead of the autumn conference, but the organizing committeeﾕs ambitious plans for this event (cf.Fig 1) gradually had to be adjusted in order to meet the financial limits imposed by a shrinking granting situation. This spilled over as a squeeze on the 2005 seminar which could only be partly alleviated by inputs from BF and its ﾓCorporate Affiliatesﾓ

 

Many factors, including the establishment of an office for the  Japanese Society for the Advancement of Science (JSPS) at KI, however underlined the need to make special efforts to find funding for a joint effort between Sweden and Japan. Also there was a need to consider an offer by the Swedish Foreign Ofice to use the Nordic Pavillion at the World Exhibition at AICHI (ﾓNatureﾕs Wisdomﾓ) for an videolinked session to Stockholm in the early autumn of 2005.

 

Against this background C.-G.Hed始 gratefully accepted an invitation to present a paper (Hed始, 2004a) at the 10th Anniversary of the ﾓZero Emission Research Initiativeﾓ (ZERI), which took place at the UN University inTokyo on Sept.16th 2004.

 

Not only did this give an opportunity for talks with representatives of the World Exhibition, Also the Swedish Ambassador and his science advisory staff kindly arranged a meeting with the President of the Science Council of Japan, Kiyoshi Kurokawa, and one of his Council members, Prof. Takao Nuki, as well as with some other leading experts.

 

On this occasion, which indicated a genuine japanese interest in extended cooperation with Sweden, special attention was given to the Councilﾕs policy document: ﾓThe Japan Perspectiveﾓ. An english edition of this remarcably farsighted document had just been published (SCJ,2004),and has since been supplemented with ﾓJapan Vision 2050ﾓ (SCJ, 2005).

 

Assuming that the Embassy discussions would revolve around long-term aspects on sustainble development, Hed始 also tabled a paper inspired by climate change issues: ﾓTowards Amphibian Habitats as a Means to Reduce Shoreline Vulnerabilitiesﾓ (Heden, 2004b). This was felt to offer opportunities for a long-term joint effort, possibly centered on Gotland which was represented by a large delegation in Tokyo, at the time of Hed始ﾕs visit. However, with regard to the 2005 seminar a focus on a pressing common concern, namely problems associated with ageing populations, seemed to deserve priority attention.

 

Taking the lessons learnt from the 2004 seminar into account, this focus of the planning was maintained and strengthened when prof.em. Lennart Levi, a senior expert in medical sociology and stress research from KI, was added to the organizing committee. In view of  Japanﾕs international leadership in the field of service robots, this early entered the planning discussions with Levi suggesting that a stimulating preparatory Internet-exchange might give rise to a lively videolinked debate on ﾓPros and Cons of Service Robots in the Care of the Elderlyﾓ.

 

However, the wide sweep of invitations for the 2004 seminar had proved both laborious and expensive, and it also attracted too many people ﾓwho already had their minds made upﾓ. This probably had a negative effect on the response that had been expected from a dozen carefully designedﾓWhat if..?ﾓ-questions, like the following which, together with its trigger comment, was posed well ahead of the 2005 media attention on a possible treath from a bird flue virus:

 

 What if the existing stock of vaccine against a new influenza pandemic is enough for only 30 % of the population?

 

          Anticipatory social innovations:

government  control of existing stock of available vaccines and drugs; priority lists / hospital staff, most vulnerable agegroups, key administrators, the public etc./; accelerated switch from eggbased to tissue culture production; preparation of manufacturing manual and storage of essential production chemicals; stand-by agreements with the pharmaceutical industry to implement immediate production switch-over dispersed production at cellbiology laboratories for geographically limited distribution of vaccines, Moratorium for unrelated ongoing research.

 

The lack of response to such trigger questions indicates a lack of willingness among experts to spend the time necessary to participate in  discussions which ﾓfall on somebody elseﾕs tableﾓ. Conseqently the What if--? approach was dropped from the 2005-plans which also, for financial reasons, came to emphasize the need for a simple communication protocol aimed at people who might be expected to have a genuine  interest in the theme selected.

 

Also lack of funding made it necessary both to abstain from serveral exchange visits to Stockholm and Tokyo, and to cancel the initial plans for a joint Visby roundtable on Sept 23rd, aimed at a discussion of the ﾓTowards Amphibian Habitats - - - ﾓ proposal. In hindsight, and considering the growing concern about the effects of global warming (the tsunami- and New Orleans catastrophies, the growing number of floodings in Europe, the increasingly violent hurricanes and the rapid melting of the polar and Greenland icecaps) this is much regretted.

 

 

6.The seminarsﾕ roots in biotechnology.

 

The choice of service robots as the major theme for the seminar has caused some comments from a few bioengineers who have felt a need for a clarification about the link between this theme and the ﾓclassicalﾓ core activities of the Bioengineering Division of the Royal Swedish Academy of Engineering Sciences.

 

To explain this it is necessary to go back to the early history of biotechnology (Bud, 1994), when many people used other concepts, such as ergonomy in order to satisfy their need for a delineation of the activities which are now practiced by ﾓbioengineersﾓ. There are now more than a dozen definitions of the subject area, but Hed始 argues for the first one, which says that biotechnology is:

 

Every application of  biological knowledge which is:

1.   Socioeconomically significant, industrially or in the

informal sector of the economy,

2.   requires transdisciplinary interactions, and

3.   involves microbiological techniques (sterilization,

microscopy, environmental control etc.).

 

ﾓBioengineersﾓ obviously need biological knowledge whenever they transfer microbiological techniques into various engineering fields (sterilization in exobiology and in the design of biomedical devices, microscopy for quality control in microelectronics, environmental control in phytotrons, sewage- and recycling plants etc.). But such transfer efforts also require an ability to extract relevant biological knowledge from many other fields such as genetics, biochemistry, biophysics, statistics and medicine. This underlines the importance of a transdisciplinary orientation. Then mathematics is often a catalyst, and sometimes the interactions result in technology clusters. such as genetic engineering. They can have such a dramatic socioeconomic impact that they take centerstage in the rainbow of biotechnologies: 

 

      ﾓREDﾓBIOTECHNOLOGY 

Pharmaceutically active products

(cancer & heart/lung drugs, hormones etc.)

  

"WHITE" BIOTECHNOLOGY

(bacteria, fungia and yeasts for bulk products and recycling)

GENETIC ENGINEERING

"GREEN" BIOTECHNOLOGY (agriculture dreived foods and forstry based biorefineries )

 

ﾓBLUEﾓ BIOTECHNOLOGY"

(foods and technical products

from marine and freshwater

resources.)

  

 

 

 

This multifaced structure of biotechnology can serve as a reminder of the fact that our efforts to guide human ecology towards ﾓsustainable developmentﾓ are limited by our planetﾕs biological diversity and by our willingness to:

 

IMITATE NATUREﾕs STRUCTURAL AFFINITIES:
                choosing processes with similar constraints,

                avoiding separation of reaction steps,

                integrating disposal techniques.

IMITATE NATUREﾕs SPATIAL DESIGNS:

                reducing transport energy,

                exploit microenvironments,

                design for change.

IMITATE NATUREﾕs FUNCTIONAL INTEGRATION

                dismantle and reuce building blocks,

                utilize ﾓwaste heatﾓ.

 

A bioengineer can draw much inspiration from such guidelines obtained from the recent World Exhibition in Japan (ﾓNatureﾕs Wisdomﾓ), but he will also see the need for advances in bioinformatics in order to meet new needs for instance in pharmacogenomics and nutrigenomics (medicines and diets tailored to an individualﾕs genetic make-up).

 

Both existing databases (literature, prevalence maps, molecular structures, medical records etc.), as well as biobanks (pollen- and microbial culture collections, tissue samples, antibodies etc.) must be expanded, and special correlation efforts must be made to support demography, epidemiology, ethnicity and gender studies.Otherwise evidence based preventive medicine and good health care will be slow to appear.

 

It is at this stage that information technology, embedded in  autonomous devices, enters the bioengineerﾕs field of vision. He for instance draws inspiration from the data coming from high speed laboratory robots and from the multitude of communication devises that are used to support all the diagnostic tools and surgical interventions that have developed greatly in the last couple of decades (Hed始 and Ill始i, 1975 a,b)

 

In particular the bioengineerﾕs focus on socioeconomic relevance draws him towards those engineering sciences which can help him to: 1/reduce risks to people and property (reduce heavy work loads and hazards, guard and provide surveillance), 2/ replace limbs or organs or to extend the limits of human perceptions impaired by handicaps and old age, or simply to 3/help society to foster political imagination.

 

The latter often illustrates our inability to fully understand exponential growth tendencies in technology (Kurzwell, 2005). This can for instance be illustrated by the length of time it takes between the invention of a device and its use by a significant part of the (US) population. The same exponential growth is also seen in communication, as expressed in the number of web-server computers on Internet. or in DNA sequence data in GeneBank. However, the most thought-provoking exponential growth is seen in computing power, expressed as calculations per second per US$ 1000. This is now around 1010 and is expected to reach 1014, around 2020. This happens to be close to the capacity of the human brain.

 

Bioengineers may not yet have fully realized the significance of this fact. not only for the intelligence of service robots but also for getting new generations of programmers to start imitating nature. faHowever some game creators are already making headway in ﾓaffective computingﾓ, when they design software that tries to recognize and align to their users emotional states (Daviss,2005). There is for instance the screen character Laura, born in MIT:s Media Lab with the ability to influence our old mammalian brain by speaking with a soothing voice, smile, raise eyebrows, make friendly hand gestures etc.

 

It has been shown that such a ﾓfriendﾓ inside instructional software can both improve language learning and act as a personal trainer in fitness programs for elderly people.

 

Even if the ﾓcyberbuddyﾕsﾓ vocabulary and stock of emotional expressions can be large, the capacity for responding to speach is normally limited and this calls for keyboard-inputs. However, the positive effect on demented ptients, which a responsive toy-animal can have (Wada et al,2005a), makes it easy to visualize a future when affective links to human cybercharacters becomes a geriatric design goal. An alternative might be to let old people experience the emotional intelligence of living animals, which means that we might then have to play down the common concern that dogs and cats cause allergies, add labor etc.

 

 

 

 

7.Comments from the Internet discussion.

 

Since the full text of the Internet discussion (Aug 1st-Sept 22nd 2005) will be available for a year (until Sept.22nd 2006) on the Biofocus Foundation homepage (<www.biofocus.org/phpBB2>), the following points are only intended to illustrate the ﾓroad mapﾓ which was used to prepare the agenda for the Sept.22nd videodialogue between Sweden and Japan. The free and simple,ﾓBulletin Boardﾓ- conferencing technique that was choosen for the exchange proved very sucessful, and the only disappointment was the dominating participation from Sweden. There was in fact only one japanes input among a total of 37 articles submitted by 33 registered users. However, all the inputs were relevant and provided a useful basis for the on-line dialogue, and also for the questions that were asked at the preparatory interview that is presented in the next chapter (8.ﾓThe Voice of Experienceﾓ).

 

From the start the Internet-discussion was limited to participants known to the fairly large organizing committee, or to some other experts familiar with the work and interests of individuals who might contribute to a useful discussion. However, it was not a closed meeting, since those approached in the first round of invitations were asked to forward the access-string to others whom they felt might be interested. The newcomers were given a few practical hints, such as advice for busy people to arrange for a secretary of a graduate student to handle their mail and to familiarize themselves with various help-sections: on registration, on limits of the size of inputs (500 words), code of conduct (no unsolicited advertisements or abusive mail),ﾓFAQﾓ (frequently asked questions) etc.

 

The conference was operated by the Biofocus Foundationﾕs project manager (Tommy Jonsson) and by two moderators who handled the scientific and political dimensions (Lennart Levi for social medicine and stress research, and Carl-Goran Hed始 for engineering aspects and aerobiology).

 

In an effort to delineate the theme of the discussion the following summary (based on a Brit.Govt.-diagram) was presented as a reminder of the fact that ﾓHealthﾓ is the product of many factors, each of them full of political ﾓPoisson-trapsﾓ, when the vulnerability of the very young and the very old are easily neglected. It was underlined that the current seminar would try to give special attention to just a few of the engineering aspects of the following

 

 

ﾓLife Qualityﾓ breakdown:

A.   Genes, Sex, Ageing

B.   Poverty, Employment, Social Exclusion

C.   Air Quality, Housing, Water Quality, Social Environment.

D.  Diet, Physical Activity, Smoking, Sexual behavior, Drugs.

E.    Education, NHS, Social Services, Transport, Leisure.

 

Much inspiration for the agenda design came from the theme of the World Exhibition ﾓNatureﾕs Wisdomﾓ at Aichi, but also through a social scientist and economist, Was始, who had recently presented his robotics thesis at the RO-Man 2005 International Conference on Service Robots (Aug.13-15). He regarded service robots as products of an incremental progress in health care development rather than as a last resort, and he emphasized the helper and companion aspects, when humans and machines  complemented each other.

 

Increased self-determination ought to be the consequence when the traditional dependence on health care providers is reduced. The RO-Man 2005 conference had illustrated the importance of transdisciplinary interactions: business administration, psychology, robotics, engineering, linguistics and cognition. Also the conference had given attention to the liability problems and insurance costs that had to be taken into account when a service robot was operated outside the laboratory.   

 

The Internet discussion should focus on such cultural and technical interphases that might inspire creative, transdisciplinary thinking about needed social and technical inventions that would fit under under the three themes that had been  selected for special attention:    

1.ﾓRobotization: Pros and Consﾓ (with a focus on service robots for the elderly),

2.ﾓImproved Indoor Air Qualityﾓ, and

3.ﾓThoughtful Habitatsﾓ (to accomodate general health care policy, architecture and the settings in which service robots and ventilation systems operate).

 

The two co-moderators started the discussion, with Levi stressing that a first step must be to determine where the relevant high-tech stands to-day, and how well service robots are accepted in different countries. He saw both a growing demand for improved care of the elderly but also pointed at a number of ethical questions which this raises. Levi asked: could it even be regarded as a human right not to be cared for by a service robot?

 

Levi illustrated the problems ahead for Europe by the following figures:

             80+          increase by 17% 2005-10

                                 increase by 57% 2010-30

                                 increase from 19M now, to 35M  2030

 

In a Green (discussion) paper from the European Commission it is noted that many widowed women live alone, often in relative poverty. Also the fertility is low and most people want to live at home as long as possible. This wish is however difficult to satisfy due to an increasing need for high-intensity care beyond 80+. Also grown-up children in the workforce may live far from their parents, who are often regarded as a burden if they canﾕt serve a useful purpose, such as  babysitting.

 

Obviously new technologies, including service robots which perform tasks that are repetitive, boring, strenous or dangerous are needed to support older (and younger) care-providers, retired persons as well as the very old.

 

A very powerful computer with equally powerful software, housed in a mobile body and able to act rationally and empathically, on its perception of the world around it, naturally makes us start thinking about the androids and humanoids that have become familiar to the theatre and movie public ever since Karel Capek (1890-1938) in 1921 coined the word: Robota (= servant, enslaved worker).

 

In a sense a merger between human beings and machines (robo sapiens) has already started to emerge in the form of advanced prosthetic devices. Considering expected advances in programming and in high-capacity hardware, it might be useful for bioengineers to keep the three laws of robotics (1938) in mind:

  1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
  2. A robot must obey orders given it by human beings, except where such orders would conflict with the First Law.
  3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

 

In 2004 those ﾒlawsﾓ were supplemented with a ﾒWorld Robot Declarationﾓ for a future when robots might have become the biggest market of the 21st century and a routine part of everyday life:

1.Next-generation robots will be partners that coexist with human beings.

2.Next-generation robots will assist human beings both physically and psychologically.

3.Next-generation robots will contribute to the realization of a safe and peaceful society.

 

So, faced with the potential of service robots Levi asked:

why?

Overburdened care-givers

 

and why not?

Levi suggested that the dialogue between Sweden and Japan should try to illustrate how the two ageing populations now try to make  socially and ethically responsible choices among a variety of new technologies that have potential to meet new needs, opportunities and risks.

 

Even if demographic factors are important, Tornstam warned that too much demographic attention might give the impression that replacing human care with robots shows a lack of empathy. An exception might be fields like assistance with personal hygiene, and where the key arguments would be quality, integrity and innovation rather than demographic pressures.

 

With regard to innovations Beck-Friis expressed the view that design engineers should use their imagination to create routines that would put them in the position (apraxi, agnosi, afasi) of the +80. The overriding goal should be simplicity, possibly competing with safety for first place. ﾒOverengineeringﾓ, as well as solutions which would tend to serve as constant reminders about cognitive limitations, should be avoided. Christensen agreed and underlined that a central design goal ought to be that the system should be driven by the user and respect his/her requirements. The safety aspects might for instance require that soft and flexible materials are used in the manufacturing process.

 

Hed始 noted that the technical frontline in robotics was quite different where industrial robots could easily be adapted to meet medical service needs, and where astronaut-like robots were considered. As an example in the first category he visualized an exoskeleton robot replacing a persistent and patient physiotherapist, perhaps working in tandem with a biofeedback electrostimulating device.

 

The technological frontline of course moves particularly fast when influenced by major funding agencies such as NASA and DoD. By replacing the mental traps of ﾒwheelﾓ and ﾒchairﾓ with ﾒlegsﾓ and ﾒsaddleﾓ they had for instance paved for way for thoughts about  ﾒstair-walking wheelchairsﾓand environmenatlly friendly power sources (batteries, fuels cells and hybrids for novel mobility systems. Lagermalm strongly argued for this field as a challenge to comprehensive city planning. Supporting mobility would help to provide the physical and mental stimulation that will become needed in order to balance an overload of passive entertainment.

 

NASA and DoD-funding has also accelerated the development of distant medical interventions for instance via surgical robots. However, Hed始 pointed out that the dramatically improved results, for instance in the case of radical prostatectomies (Fig.2), that has been achieved by prof. Peter Wiklund and his colleagues with a very large and expensive robot at the Karolinska Hospital (fig.3), cannot yet be fully exploited. The reason – which is an important science-policy memento – is that such partnerships between man and machine require demanding educational efforts. Also the full power of this  partnership will have to wait for supporting laboratory robots, equipped with highly specific and sensitive diagnostic tools (such as guest Wingeﾕs proximity ligation), that are now on the horizon.

 

Hed始 also bridged the gap between ubiquitous intelligence in hospitals and in homes (ﾒThoughtful Habitatsﾓ) by making  some comments about ﾒsmart housesﾓ with all their electronics embedded out of sight (in household aids, entertainment and communication devices as well as in systems that control temperatures and ventilation).

 

He found it easy to visualize how service robots can expand their own intelligence by interacting with such ubiquitous intelligence. As demonstrated by Fujii, a smart house could for instance have a bathroom with equipment that - without physical contact with a patient – can pick up an ECG and other signals that might alert a service robot and/or a nurse or doctor.

 

If one excepts fire-fighting in ﾒsmart housesﾓ, low-cost environmental control (temp,humidity), is easy to achieve, and as pointed out by Jonsson firefighting need not be a major hurdle if it is taken into account at an early planning stage and uses warning signals adapted  to the clientsﾕ levels of cognitive impairement.

 

However, air quality can be a problem for frail people as shown in Hed始ﾕs introductory paper for the Air Quality Section: ﾒMicrobial Aerosols as a Health Hazard for the Elderlyﾓ. This paper, which originally was focused on September 22nd as ﾒthe International Car Free Dayﾓ supplemented a presentation the previous evening. Then Landahl from the Stockholm Environmental Office, described various efforts to improve the cityﾕs air quality.

 

However, the focus of Hed始ﾕs paper was on personal hygiene  (showering, hot tubs), as well as on rehabilitation pools, all considered in an aerosol perspective. This brought both Legionella spp., and a number of other fastidious microorganisms into focus, in particular because some of them produce potent hydrophobic toxins. According to Solvicon (an international environmental technology company), the Legionella situation can be monitored with a quantified polymerase reaction (QPCR), and also the toxin producers can now-a-days be characterized genetically.

 

However, since the toxin-producers attack mitocondrial membranes which might be an important factor in the ﾒsick house syndromeﾓ, Hed始 highlighted the immobilization of boar sperm, a subject selected for presentation in the afternnon by Salkinoja-Salonen, as an ideal screening tool. Hamer then also would elaborate on his Internet points. There he mentioned UV, which has no residual effect, high temperature, which does not protect against pathogens in diluting cold water, and detergents, where counterdiffusion of nutrients in biofilms needs to be taken into account. This in fact constitutes a logical future target for BF:s international activities.

 

Nilsson  regretted the geographical limitations of the seminar, and he cited Polen as an example of how strong traditions make the working generation take care of the elderly. He argued for a ﾒHuman Integrity Limitﾓ and asked if it would not be more dignified to be served by a trained monkey than by a robot. After all, a mutual emotioinal stimulation (a dog leading a blind or picking up the morning newspaper) should be regarded as an asset.

 

Also Christensen brought up the international aspects of service robot development. He for instance mentioned South Koreaﾕs ten national economic development programs, each with a budget of $ 100 mill. per year over a ten year period. One of the areas which has the aim of being the 3rd largest industry provider by 2015,  divides one of those sums equally between industrial robots (for the car industry) and service robots (for industries like Samsung. LGe and Yujin, also acting as guides towards a small scale service robotics industry).

 

Considering such ambitions and considering also some 60 prototypes that were presented at The World Exhibition in Japan last summer, a major need will obviously be the technical evaluation of various designs. This should be coupled with a cost/benefit analysis of their performance in various markets such as Europe. Here the engineering research has for instance lead to designs like the Care-O-Bot from the Frauenhofer Institute for Production and Automation). Obviously such initiatives can be expected to stimulate industries like Siemens and Phillips (Aware Homes and Adaptive Technologies) to integrate existing technologies in  ﾒsystemsﾓ  that can cover long distances, provide physical and cognition support, assist in finding locations, strech for, fetch and carry objects, reduce stress and accidents in the work-place etc.etc. The recurring question will then be: what will they cost and how many ﾒclientsﾓ can those systems serve in one location?

 

To answer such questions will require a test setting that is representative for the national public health system (PHS). This normally operates at different levels (state, regional, municipal and local), and with costs and responsibilities divided between an insurance-, a governement- and a private sector.

 

All those components have a long history of exposure to strong political and union pressures, which means that adoption of new technologies tends to be a slow process. As illustrated by the early adoption of different IT-solution in some major Swedish hospitals procurement can also prove to be very costly and unpopular when standardization becomes inevitable. Then the ambitions are often very high and involve a safe and comprehensive medical record for every citizen.

 

As a stepping stone towards such a major system, perhaps serving as an EU-model,  Akner suggests a focus on the elderly with an effort to correlate their multi-morbidities with their functional impairements. This is a field which he regards as under-researched, under-diagnosed and under-monitored, particularly in the big cities, where staff circulation at the primary helth care level makes it hard to follow the ageing process over time.

 

As a consequence Akner argues for an evidence-based best practice via a replacement of text-based diary records with graphic representations of changes over time (laboratory data, weight, functional rating scales, severity scores for criteria-based diagnosis etc.). This DTF-system (=Diagnosis, Treatment and Follow-up) then naturally evolves into a training tool (PCS=Pedagogic Communication System) not only for professional c but also for the patients and for their relatives

 

 

8.The Voice of experience, interview: Joseph Engelberger.

 

The structure of the videodialogue between Stockholm and Tokyo was based on impressions gained from the preparatory Internet discussion and a starting ﾓhistory of scienceﾓ-input. This had the form of the following telephone interview on Sept. 14 between co-moderator Hed始 (C.-G.H.) and the still very dynamic 80-year old Joseph F. Engelberger (J.E.), who once introduced robots on the scene of industrial productivity.

 

C.-G.H.: Good morning, Dr.Engelberger, how are you, up bright and early. I wonder if you have had a chance to look at the E-mail I sent you yesterday?

 

J.E: Yes, the one with the questions?

 

C.-G.H.: Yes, thatﾕs right. I sent them because I had a feeling that they were a bit hefty, and that you might like to delete some and concentrate on others.

 

J.E.: Yes. I have looked at the questions, but I didnﾕt get the attachment on the program, because my agressive virus protection cut it out. However, I now do have all the questions in front of me, I will take them anyway you want them. Are you taping this?

 

C.-G.H.: Yes we record the discussion and will  send you a disc for your records in due course. If the questions are all right with you, I will now start with a brief introduction for the benefit of the conference on Sept.22nd.

 

At the joint Academy Seminar, where we will try to get a grip on the pros and cons of robots in support of the elderly, it is a very great previlege indeed for us to get this chance to draw a little on your very long experience in the robotics field, Dr.Engelberger.

You were the first to see the potential of industrial robots, and like many other successful entrepreneurs, a garage saw you join with the inventor, George Deval, and then create the ground-breaking company UNIMATE. In the early sixties the success of your robots in a General Motors plant showed how acute your commercial acumen was, when you predicted the proliferation which then followed right up to, and through the Second World War.

However, in the course of the last few years you have repeatedly expressed great disappointment in the fact that industry has failed to translate so little of this huge fund of experience into cheap and multifunctional service robots in the homes of a growing population of old, and often mentally handicapped people.

The pressure on industry which you have maintained over the years, not least through the prestigeous Joseph Engelberg award, is most admirable, but it still does not seem to be the R&D departments which push robotics, but rather the advertising consultants, who like to add the word robot to pretty dumb gadgets.

I will now ask you a few questions which lead up to the last one which reads: suppose you would launch a major program to develop an affordable service robot for the old and mentally handicapped. Would you develop it under the umbrella of a single industrialized nation (US, Japan, S. Korea, UK, Germany, France, Italy, the Scandinavian countries etc.). Or would you aim for a true chimera from the start (design from Japan and the US, communi-cation technology from Scandinavia, programming from India and large scale manufacturing from China etc.)?

 

Now comes my first question:

 

You actually advocated ﾓservice roboticsﾓ as early as 1989, when your  book ﾓRobotics in serviceﾓ was publishedﾓ. How come that we to-day only find rare demonstration robots in the homes of  elderly people, even in such a highly industrialized country as the US?

Is there a lack of a vision such as the ﾓHumanoid Programﾓ in Japan, a lack of political will to define an agressive national goal (going beyond the ﾓmore of the sameﾓ business approach), as in South Korea, or is there perhaps a lack of focused funding as in Europe, which however recently approved 130 mill. ﾛ to integrate industrial and service robotics. What do you think?

 

J.E.: Well. prof. Hed始. I donﾕt think that it is any of those things. What is wrong is not lack of funding. My estimate is that a fully appropriate, able, caregiving robot would take development costs, not research costs - but development costs , of about three million dollars and take 27 months to do it. I would be very much against trying to get a consortium of countries together. That has been tried in the UN. My feeling on it is that it is not such a big deal. Any of the robot manufacturers, and indeed any of the companies that provide care for elderly people, could run this program. Now the difficulty is to convince them that it can be done. It is very hard. People think that it canﾕt be done. It can be!

 

C.-G.H. You sounded a very optimistic tone just now.

 

J.E. I am optimistic. Well, let me say in Sweden I know that P.G. Hedstr嗄 would like to do this, but he cannot get his management  behind it. ABB, who licens Unimation,, is very rational and logical to do the thing. The trouble is that to-day management looks for faster returns and short brains.

 

C.-G.H. Well, that is a great problem almost everywhere, istnﾕt it?

 

J.E. So, wether it is the United States or South Korea. You know, I have been to every place, all the countries. talked with leading companies in all of them, and I say: I donﾕt need a company behind it. Iﾕll tell you another thing: When we finally started Unimation, I had talked with 46 different companies to get sponsorship and finally didnﾕt get it from a logical one at all, I got it from a railroad car company. Not from IBM, United Technology, General Electric, General Motors. Every one of those companies, after they saw us getting into the business, started themselves.

 

C.-G.H: Istnﾕt that then a sign of lacking vision at top management?

 

J.E.: Itﾕs top management I think. They are big and smart and smarter than a little jerk in Connecticut, and they went their own way, they didnﾕt do the marked studies that we did, and they didnﾕt understand something which is very important, and that is cost/benefit. We will talk more about cost/benefit later.

 

C.-G.H. Then I come to my second question: You have pointed out that many of the technologies needed to make a service robot useful to the elderly are now available ﾓon the shelfﾓ (artificial intelligence, speech recognition, machine vision and virtual touch etc.). However,  the industrial robot makers, for the last twenty years you say - at least in the US – have failed to put the bits and pieces together in order to face the elderly as a very large and growing market. What holds them back?  Could it be a dense forest of patents and a multitude of hungry lawyers, who might be less of a problem in Japan, South Korea and Europe, than they are in the US.?

 

J.E.: Well IﾕlI tell you: I join you in concern about agressive lawyers, but this is not the issue.There are no patent considerations, the technology has been disclosed. If, for example, there was a patent on voice recognition, well, all we would do is to license that technology, if there would be something about vision, we would buy that vision package. There is no barrier at all from patents. I think the big thing is that people donﾕt believe it can happen. Let me take one thing; speach recognition. I had an opportunity to make a presentation to

Ross Perot, the wealthy industrialist. When I went to see him in Texas he had just had a skiing accident, and he had to take his broken leg to a doctor, so he left me with two liberal arts executives. They listened politely and one said: I donﾕt think you can possibly do it. We know, because William Gates has spent millions of dollars just trying to have a computer understand syntax. What on earth has syntax got to do with anything? We donﾕt need to have great english

or great swedish or any great language to deal with a robot.

If I tell a robot: please go to the refrigerator and get me a beer, or please get me a beer, I just say Beer!. The robot is going to know what I want, and it knows it in something like we did in the islands of the Pacific. We effectively enslaved the brown people there, and we worked up a language for dealing with them. It was called pidgin english. How about pidgin robotics?  There isﾕnt any barrier for communication - voice recognition is so advanced today, as long as you realize that it has to be speaker specific. If it is going to talk with anybody on the telephone. it must be speaker specific.

 

C.-G.H.: Here was again a note of optimism, so here comes my third question:

 In an ordinary modern apartment we take the intelligence, which is built into our communication-and entertainment equipment, as well as in the dishwasher, washing machine, microwave-oven and refrigerator, for granted. It is when a computor is introduced to monitor and control all this, plus a number of light-, temperature-and humidity sensors, that we talk about ﾓsmart housesﾓ. Since the industry is busy improving all those different items, isnﾕt the ﾓSmart Houseﾓ actually a strong competitor to the cheap, universal, service robot?

 

J.E.: Really, not at all! Smart Houses is a wonderful technology and very useful for people who are elderly or infirm, or just to everyone  of us. But just remember what the smart house does: it communicates with all of those different things, lights on and off and so forth, but it does not catch and carry. So even if you can communicate from the computer to the microwave and have it do something, no one gets it and delivers it. No one is cleaning the smart house – it just doesnﾕt do things. It is a great adjunct, and in many ways you would want it, but on the other hand the robot is going to be able to speak with wide eyes to any piece of equipment in the house. Thatﾕs straight forward, the robot will be able to purchase, to do shopping by telephone. Then it will help a person walk, perambulation. Smart houses donﾕt help ambulation. It will help a person to get out of bed in the morning. Elderly people in the United States, some 300.000 older people, fall and break a hip each year. If you are 65 years old they put you together again. If you are 85 years old, you will never walk again.

 I have talked with people in the visiting nurses association, and they tell me: look, sometimes when we go to a home with a caregiver in it, the most important thing is to let the caregiver go outside and scream. Older people are a pain in the neck, and they are a strain on a caregiver. A robot on the other hand can take more scum, it has a thick skin. Abuse it all you want verbally, and it wonﾕt mind. Smart houses are not a competitive thing at all, it is an add on.

Let me tell you something  else, if you read science fiction at all. There are two approaches, One was Ray Bradburyﾕs – a famous writer - he had a story about a smart house. The house was a robot, it actually physically moved things around, physically cooked meals, brought the dishes to the table, took them away, broke up the dishes and smashed them and built new dishes, and kept working. After a nuclear war the people were dead, but the house kept working. Asimovﾕs  approach was: no,no!  –make the robot one of us.

 

C.-G.H. Well you have a point there, because many people associate so directly to science fiction, that they cannot visualize things which fit a future which is not very distant.

 

J.E. Iﾕll tell you, I think that Asimov was right. He knew what was going on - he went to Columbia University some years before me.

When I read his I-Robot stories I said: My Gosh - to do something like that! That was a sort of motivation, and from that point on, when I picked up the right George Duvalﾕs patent - which was not a robot but a programmable manipulator - I saw in it the potential. There was a patent which was good and which we ultimately licensed to ASEA and to General Electric, IBM, Cincinnati Milacron, but since that time all the patents have run out that are fundamental to robotics.

 

C.-G.H.: Well, letﾕs then get on to the question of acceptance. The

value of a dialogue between two distant countries, like for instance Sweden and Japan, is that it might illustrate differences in culture and traditions which influence the market penetration of new high-tech products. This is particularly true, I think, in case they touch upon the display of emotions and on ﾓpersonal integrityﾓ.

 For all I know, old people in Sweden might see robots as more treathening, than they do in the United States. This could mean that interactive toys like a talking seal (PARO), a dog (AIBO) or a teddy bear companion, might be more important for widespread acceptance of service robots in Sweden, than in Japan.

 

When the mental facilities fail, I would guess that the sense of control over a service robot is important. You might for instance choose the different programs that go into the computer which controls specific household tasks (for instance Haipeng Xieﾕs Dr.Robot designs), or you could simply ﾓpull the plugﾓ. Do you think that widespread acceptance by elderly people might actually come easier for wheeled personal assistants than for walking humanoids?

 

J.E.: Well, I am dead set against walking, and you want to discuss that later. They have been expressed I think, but I think that walking was a mistake. Walking is a trerribly expensive adjunct to a robot because elderly people    infirm and elderly people - should live on one floor. In navigation on one floor, or many floors, to help make robots navigate through sixteen floors of a hospital, knows where every department is, knows where every nurses station is, doesnﾕt bump into people, doesnﾕt hurt anybody and does its job. Itﾕs a given that you can navigate.

And now the next thing is: how do people react? Itﾕs amazing with this big box on wheels. But it talks: my way is blocked, will you please remove the obstacle? People stand in front of it, and they love it! People are so encouraged. I must admit that my experience is all in the United States. But it was surprising to see how quickly people anthropomorphized a box on wheels, because it talks and moves independently.

And go the other way: if you have a Robosapiens or a Robovac, see this issue I mentioned before is cost/benefit. Itﾕs not just the cost, itﾕs the benefit. Now, why is robovac so sucessful and robosapiens so sucessful: itﾕs cheap and the cost of a hundred dollars for a Robosapiens, 200 dollars for a Robovac, is small compared to the fun, the sheer fun, you get out of it. Robovac doesﾕnt work worth a damn, and by the way, if you look at the distinction: we did work for Electrolux, they came out with a vacuumcleaner called Trilobite, 1500 dollars.

 

C.-G.H.: I have seen it, and I thought that it was more expensive than that, and I think that they have competition from less expensive units.

 

J.E.: Well see, thatﾕ s the point. I called the head of Robovac, I tease him and I say that Robovac is the perfect vacuumcleaner for a round room with a wooden floor and no furniture. But out of Electrolux itself I learned that you need 3/4 hp to really clean a rug. And in the Unites States, where they sell many vacuumcleaners, the average vacuumcleaner is used less than 100 hours per year. So if thatﾕs all the use itﾕs going to get, you got to have a cheap one, so that you can give it to your mother-in-law for Christmas., and have a drink while. it scopes around, and then leave it in the closet along with the Herorobot and a number of other old ones that died.

 

C.-G.H: Well, I entirely agree: the pricetag is crucial. Now, that is what makes me wonder weather there is really a market ever for bipeds like Hondaﾕs ASIMO ($ 390.000 dollars in 2004), or for SONYﾕ s QRIO (450.000 dollars). This certainly tends to limit their usefulness. I guess that it is fine for high-tech education or for status jobs in major corporations. Of course, there is a lot of intelligence which has to go into keeping balance, particularly in rising up after falling, and basically to move around.

I agree with you entirely that one could do a lot more about moving. I have for instance been interested in looking at ﾓwalking wheelchairsﾓ. There might be particular instances when one could advocate them, but then I think that you would also have to pay a bit more. But you also mentioned those hospital robots that can follow the rounds of a doctor. I have seen pictures of one that has been tested at Harvard Medical School and does just that. It letﾕs the doctor keep in touch with his patients without really being physically present. It is said that this is very well accepted by the patients. Now, that is of course a very limited function, but with simple things like delivering food  or drugs and so on, could be pretty simple. The price for such a robot could probably come down dramatically. Now, the question is just how can one pass the number-barriers around, so that one can bring the prices down.

 

J.E.: Well , the biggest issue in my mind is the price issue. At  some point someone has got to recognize it is not the cost, it is the ratio of  cost to benefit. Now, if you use a robot in conjunction with professional medical people you can help them. They are very high-tech people,  and if you can reduce the cost of their service you have got something. On the other hand, if all you are going to do is to deliver food (and donﾕt think that that is so simple!) it is different. A robot that helps in minimal invasive surgery for example, is quite sophisticated, and so is a robot that roams around a hospital delivering food and other goods 24 hours a-day, seven days a week. It replaced human runners that make 12 to 15 dollars an hour. The robot on a rental basis, which is what we did with HelpMate, runs about 5,50 an hour, and it also allows the nurses to stay with their patients, where they belong, in stead of running around the hospital. The cost/benefit does not justify the prices of ASIMO.

 

C.-G.H.: This is one of the reasons why we have brought robot assisted surgery into the picture, even if it is a question if you should really talk about a robot, since it is the surgeon who does the job, but the cost/benefit of those machines is of course noteworthy.

 

J.E.: But ﾓcost/benefitﾓ is wrong! There is no cost/benefit for ASIMO It is a cute design, and it takes great engineering efforts to make it walk. But walking is unnecessary. Why 390.000 dollars. You know, 27 of the articulations in ASIMO are all in the legs and need control. You donﾕt need them for a rolling robot. It takes two articulations only. And the next thing. ASIMO doesﾕnt do anything. It walks up and down stairs - big deal!. You should live on one floor if you are old and infirm. If not, and if you want to go upstairs - have an upstairs maid - then you have one of those elevators, that runs right alongside the stairs, they have for older people and cost about 1500 dollars.

 

C.-G.H.: Well, I know. The reason we brought up robot assited surgery

was the fact that the cost/benefit of having an exceedingly steady hand and good 3D vision, is really what counts.

 

J.E.: Let me give you an example from Unimation days. When we first started to sell Unimate, we were very interested in cost/benefit. The benefit has to be there, so the price was set at 25.000 dollars for an industrial robot. They cost us 60.000 dollars to manufacture. That was not good business, so our investors said: what can we do about that? They brought in McKinsey manufacturing consultants. and they looked at our drawings and they said: look, if you people do some value engineering, some tooling, and if you bring the production rate up to 200 per month, you can manufacture those robots for 8000 dollars. That looks good, but itﾕs like saying: gee whiz, a RollsRoyce costs 350.000 dollars - who the hell is ever going to buy a car? They are not, they are going to buy a Volvo, a Volkswagen or a Chevrolet which are produced in quantity.

I tell you: if you look at the designs we have for a rolling robot with two arms and the ability to cook and clean, fetch and carry, handle security and even carry out reasonable conversation. We estimated that it would probably sell at the same pri