I-DRESS

The main objective of the project is to develop a system that will provide proactive assistance with dressing to disabled users or users such as high-risk health-care workers, whose physical contact with the garments must be limited during dressing to avoid contamination. The proposed robotic system consists of two highly dexterous robotic arms, sensors for multi-modal human-robot interaction and safety features.
The system will comprise three major components, each of radical impact to the field of assistive service robotics: (a) intelligent algorithms for user and garment detection and tracking, specifically designed for close and physical human-robot interaction, (b) cognitive functions based on the multi-modal user input, environment modelling and safety, allowing the robot to decide when and how to assist the user, and (c) advanced user interface that facilitates intuitive and safe physical and cognitive interaction for support in dressing. The consortium consisting of three partners provides the expertise for the main lines of research required by the project: CSIC-UPC will work on perception and human-robot interaction, IDIAP will contribute to robot learning, and UWE-BRL will provide the expertise in safety and interface design.
The developed interactive system will be integrated on commercial WAM robotic arms and validated through experimentation with users and human factor analysis in two assistive-dressing scenarios. Additionally, developed robot safety features and the learning by demonstration algorithms will be implemented on a Baxter robot, thus ensuring general applicability and easier acceptance of the project results by both industry and scientific community.

Robot safety is the key issue that must be solved if robots are to leave the research labs and be allowed to enter all areas of society. A lack of widely adopted standards for personal care and medical robots is currently being considered by special working groups within the ISO - ISO TC 184/SC 2 and IEC SC62A. A new ISO 13482 safety standard for personal care robots was published in 2013; these types of robots involve close robot-human interaction for providing personal services. Dr Dogramadzi is actively involved in these working groups and the results of I-DRESS will contribute to further developments and considerations of the standard committees. Leaving the robot system in the homes with users is beyond the scope of this project, due to health and safety governance procedures. However, we will attempt to test the system with healthy adults in the laboratory environment.
The use of robot learning from demonstration techniques will allow the developed system to automatically adapt to the specific needs of each user (which can be changing with time). This enlarges the scope of possible applications as well as the potential markets, since the dressing assistant can learn rapidly what it is expected to do, without requiring the manual reprogramming of the robot through a computer language. This versatility is expected to be crucial for such challenging physical human-robot interaction.
The technological and HRI advancements made as part of the proposed scenarios can be extended to use of similar robot configurations to support other activities of daily living. The ensuing independence for individuals, previously reliant on carers and others, will mean greater autonomy to participate in society and the workplace. Even in cases where the number of carers is reduced from two to one for personal care assistance, there will be a significant healthcare saving and more effective use of precious human resources.
The impact of the human factors and interface design research proposed here is of crucial importance in facilitating understanding of mixed initiative human-robot interaction multi-modal dialogue. Together with extending application to people with accessibility needs where one or more communication modalities might be impaired, this research will produce pragmatic solutions for the use of robotics in the assistive care sector.

Final project developments will be done under the Open Source policies, and the use of the ROS tools will ensure their reusability. The aim of the project is to produce transferable skills, that once developed can be used as a module by other research groups or in industry. The dissemination of project results and strategies for contacting with various stakeholders is described in the section 3.2.