Stroke is an age-related illness with significant individual and societal impacts. The long term impacts associated with many strokes can be mitigated with timely rehabilitation. Therapy robots have been introduced to these programs in an effort to reduce the economic burden to society and to improve the level of care provided to stroke survivors. The purpose of this thesis is to develop a teaching mode for an upper limb therapy robot. The system will allow physiotherapists to interact with the therapy robot without the need for any specialized industrial training. At the same, the system will reduce the data associated with patient movements to reduce requirements for robot safety and motion systems. The proposed system was successfully confirmed using a laboratory scale industrial robot and a standalone motion control system consisting of commercially available AC servo motors and a motion controller with both generated and recorded paths.

京都大学 / 新制・課程博士 / 博士(人間健康科学) / 甲第23830号 / 人健博第101号 / 新制||人健||7(附属図書館) / 京都大学大学院医学研究科人間健康科学系専攻 / (主査)教授 青山 朋樹, 教授 市橋 則明, 教授 宮本 享 / 学位規則第4条第1項該当 / Doctor of Human Health Sciences / Kyoto University / DFAM

Les systèmes s’adaptant à leur environnement et en interaction physique avec l’homme se développent de plus en plus dans des domaines comme le médical, l’assistance aux personnes ou le travail en usine. Ils diffèrent des systèmes classiques par leur capacité à s’adapter à l’environnement et à prendre des décisions en tenant compte de leur perception de l’environnement et notamment de l’homme. La défaillance de tels systèmes pouvant avoir des conséquences catastrophiques sur l’homme, l’analyse et la démonstration du niveau de confiance que l’ont peut leur accorder vis-à-vis de la sécurité-innocuité, et a fortiori leur certification, constituent aujourd’hui un vrai défi. La construction d’argumentaire de sécurité (ou dossier de sécurité, ou safety case), est un des moyens permettant de préparer la certification de tels systèmes. Il s’agit principalement de justifier pour chaque danger comment il a été traité et ramené à un niveau acceptable. Malheureusement, dans le cas des systèmes robotiques, de nombreuses incertitudes subsistent, et il n’existe pas à l’heure actuelle de méthode systématique permettant la construction de tels dossiers de sécurité et la démonstration du niveau de confiance sous-jacent. L’objectif des travaux est de contribuer à la définition d’une telle méthode en partant d’une technique d’analyse du risque dédiée à l’analyse des interactions humain-robot, puis en s’appuyant sur des modèles formalisés de construire l’argumentaire de sécurité et d’évaluer automatiquement le niveau de confiance dans cet argumentaire. / Robotic systems that continuously adapt to their environment and physically interact with human are increasingly used in various fields like personal assistance or factory work. They are characterised by their ability to adapt to the environment, to take decision in the light of their perception of the environment and particularly of the human. As the failure of such systems may lead to catastrophic consequences, analysis and justification of the level of confidence in these systems with regards to safety, and furthermore their certification is a real challenge. The construction of a Safety Case is one of the means that can be used to support the certification of such systems. It is aimed at describing and justifying how every hazard has been mitigated and its severity maintained as low as reasonably possible. However, for robotic systems that have to deal with many uncertainties, there is a lack of a systematic approach to support the construction of their Safety Case and the assessment of its underlying confidence. Our research aims at contributing to the development of such a systematic approach starting with a risk analysis focusing on human-robot interactions, followed by Safety Case construction from formalized models and finally an automatic assessment of the confidence in safety argumentation. As a case study, the safety of a rehabilitation robot for strolling is analysed and justified based on the approaches developed in this thesis.

Advanced devices that can assist the therapists to offer rehabilitation are in high demand with the growing rehabilitation needs. The primary requirement from such rehabilitative devices is to reduce the therapist monitoring time. If the training device can autonomously adapt to the performance of the user, it can make the rehabilitation partly self-manageable. Therefore the main goal of our research is to investigate how to make a rehabilitation system more adaptable. The strategy we followed to augment the adaptability of the GENTLE/A robotic system was to (i) identify the parameters that inform about the contribution of the user/robot during a human-robot interaction session and (ii) use these parameters as performance indicators to adapt the system. Three main studies were conducted with healthy participants during the course of this PhD. The first study identified that the difference between the position coordinates recorded by the robot and the reference trajectory position coordinates indicated the leading/lagging status of the user with respect to the robot. Using the leadlag model we proposed two strategies to enhance the adaptability of the system. The first adaptability strategy tuned the performance time to suit the user’s requirements (second study). The second adaptability strategy tuned the task difficulty level based on the user’s leading or lagging status (third study). In summary the research undertaken during this PhD successfully enhanced the adaptability of the GENTLE/A system. The adaptability strategies evaluated were designed to suit various stages of recovery. Apart from potential use for remote assessment of patients, the work presented in this thesis is applicable in many areas of human-robot interaction research where a robot and human are involved in physical interaction.