3. ROBOT DEPENDABILITY IN HUMAN ENVIRONMENTS
3. ROBOT DEPENDABILITY IN HUMAN ENVIRONMENTS
3.1. Context and perspectives
A central direction of research today concerns the field of Human Friendly Robotics also noted as Human Centered Robotics. Central to this large spectrum of novel applications, the domain of Assistive and Personal Robotics, both by its technical aspects and societal and economical impacts, opens a true Grand Challenge.
In the context of IARP, this offers the opportunity to promote a vast spectrum of frontline research themes such as learning, decisional autonomy functionalities and architectures, multi-sensory perception, multi-modal human machine interfaces, . Their embodiment in integrated systems opens a perspective for efficient and realistic development. Moreover, the necessity to have the integrated system operating in contact with humans brings to the frontline of the basic research activity themes that are all too often ignored: safety, reliability, operational robustness, maintainability, . which can be compounded in a more general concept of Dependability.
All around the world growing interest and increased efforts are shown in basic research projects as well as, in several large scale cooperative programmes aiming to demonstrative realizations.
France is actively promoting contributions in this field. A major effort has been devoted to the organization in the frame of IARP and jointly with IEEE/RAS of two dedicated workshops.
3.2. A second WS framework
Workshop objectives
Considering the similarity of interest and objectives, the International Advanced Robotics Programme (IARP) and the IEEE-Robotics and Automation Society (IEEE-RAS) have signed a Memorandum of Understanding agreeing to technical cooperation to foster advances in the fields of Robotics and Automation.
Both organizations recognize the importance, within the broad domain of Human Centered Robotics, of current frontline technical issues and developments in Service, Assistive and Personal Robotics where machines have to closely interact with humans. The R&D directions emphasize in particular the human-machine interaction aspects where the person may be either a non-professional user or a by-stander, or both.
This clearly points out to the critical questions of physical safety and the multiple components of operating robustness. Both aspects can be captured by the concept of Dependability.
Unlike the industrial robotics domain where the work space of machines and humans can be segmented, service and personal robots cannot but have contact interaction. The safety aspect entailed is, of course, already an important challenge addressed to Robotics research.
Still we believe a more difficult and far reaching challenge concerns Operating Robustness. Here, the issues are central to the very concept of "Intelligent" robots.
Indeed Machine Intelligence is a necessity as soon as we consider applications not strictly related to a sole and very simple task. Task diversity in not completely engineered environments and in the presence of non-professional users, implies necessarily significant levels of Robot Autonomy and sophisticated, efficient, robust, friendly machine- intelligence interface.
A first IARP-IEEE/RAS workshop was held in Seoul, Korea, on May 21-22, 2001. The final programme complied with the central technical and operational objectives of the workshop:
A post-workshop report can be accessed on: http://www.laas.fr/rdhe/
The second IARP-IEEE/RAS joint workshop on this broad theme has been organized and hosted by LAAS-CNRS, Toulouse, France, on October 7-8, 2002.
In keeping with the general objective to foster research activity and international cooperation, and to consider mid term and long term market product perspectives, the emphasis is put on main technical issues, research directions and in relevant current work. Application cases highlight specific thematic aspects and bring in state-of-the-art know-how and best practice experience in closely related domains.
Workshop organization
The workshop was organized along sessions on research directions and on specific case studies.
Main research directions:
Special case studies were presented by designers and users in the following domains:
All sessions were set to allow for in-depth open discussion. Two thematic panels completing special sessions further provided room for discussion and assessment. A final general round table concluded the two workshop days. It summarized the workshop contributions and recommendations that will be included in a written workshop report.
The proceedings of the workshop can be assessed on:
http://www.laas.fr/drhe02/preprints.pdfWorkshop sponsorship
An important feature of this workshop concerns the joint sponsorship of the IEEE Robotics and Automation Society and, of course, of IARP. Special worth to mention is the large IARP support since besides France, which hosts the workshop, ten other IARP countries co-sponsor it with the respective country representatives engaged in its direct support: Australia, Canada, France Germany, Italy, Japan, Korea, Russia, Spain, U.K., USA.