High level functions for the intuitive use of an assistive robot Olivier Lebec 1 , Mohamed Walid Ben Ghezala 1 , Violaine Leynart 2 , Isabelle Laffont 3 , Charles Fattal 2 , Laurence Devilliers 4 , Clément Chastagnol 4 , Jean-Claude Martin 4 , Youcef Mezouar 5 , Hermanth Korrapatti 5 , Vincent Dupourqué 6 , and Christophe Leroux 1 1 CEA LIST, 18 Route du Panorama, BP6, F-92265 Fontenay aux Roses. firstname.lastname@cea.fr 2 Association APPROCHE, Montpellier, France, delegation@approche-asso.com 3 CHU Montpellier, Montpellier, France, i-laffont@chu-montpellier.fr 4 LIMSI, B.P. 133, 91403 Orsay CEDEX, France, devil@limsi.fr; cchastag@limsi.fr 5 Institut Pascal, 24, avenue des Landais BP 80026 63171 Aubière Cedex, France, fistname.lastname@univ-bpclermont.fr 5 Robosoft, Bidard, France, vincent.dupourque@robosoft.com Abstract— This document presents the research project ARMEN (Assistive Robotics to Maintain Elderly People in a Natural environment), aimed at the development of a user friendly robot with advanced functions for assistance to elderly or disabled persons at home. Focus is given to the robot SAM (Smart Autonomous Majordomo) and its new features of navigation, manipulation, object recognition, and knowledge representation developed for the intuitive supervision of the robot. The results of the technical evaluations show the value and potential of these functions for practical applications. The paper also documents the details of the clinical evaluations carried out with elderly and disabled persons in a therapeutic setting to validate the project. Keywords— assistive robotics, evaluation protocol, mobile manipulation, elderly and disabled people, intuitive HMI, empathy emotion understanding, object recognition. I. INTRODUCTION There are approximately 69.2 million people over 80 years old in the world today. This number is expected to reach 379 million by 2050 [18]. This demographic trend is an incentive for policymakers to promote the development of new services, such as robotics, to help elderly people in their everyday life. Such services would include improving their safety on public transportation or at home, the improvement of their health and wellbeing and the facilitation of social inclusion. The objective of “ARMEN” is to develop an assistive robot providing advanced functions to help maintain elderly or disabled people at home. The assistive robot SAM aims to compensate for caretaker’s limited availability. We want to give an opportunity for caretakers to restructure their schedule to concentrate on patients’ needs. In order to offer a robot that meets the expectations of users and caretakers, ARMEN has focused on the following issues: • The reliability of the mobility of the robot in 3D in an indoor environment, • The usability and intuitiveness of the Human Robot Dialog by using 1) semantic analysis of images, 2) an avatar, 3) emotion understanding from speech analysis, 4) a semantic level knowledge representation, 5) the design of various robot behaviors, • The development of several automatic functions of mobile manipulation to assist the user and the caretakers such as “find a lost object”, “bring it back” or “manipulate object”, • The proof of concept with patients under medical control, • The development of a prototype that has the potential of becoming an industry product. One of the main efforts of the work presented here has been to facilitate the usage of a highly technical device by a non-specialist. To reach this goal, we developed a set of functions to make the robot autonomous. We used a semantic level analysis to allow a dialogue between the patient and the machine with familiar terms (non-technical). We also developed an efficient supervisor with an intuitive Human Machine Interface (HMI). We first present the state of the art of assistive robotics. The previous project ANSO, which motivated us to undertake the ARMEN project, is presented in the second part. The third and the fourth parts are focused on the robot and its services. We then describe the technical and clinical evaluation procedures before briefly describing future projects. II. RELATED WORK A. State of the art Assistive robotics is a field of research that has been getting an increasing amount of attention since the early 1970’s. In this document, we refer to assistive robots as robots which are designed to either maintain an independent lifestyle or improve the quality of life of elderly, disabled and possibly non-disabled people. Research projects in assistive robotics are very diverse. Here we differentiate robot systems by their capacity to manipulate objects or not. ARMEN’s objectives are distinct from those of companion robots (such as PARO [12]), which mainly focus on emotional interaction between the machine and the user and on the usage of robots for social purposes such as cognitive stimulation, and social networking. Mobile robots with no manipulation capacities offer services such as: 2013 IEEE International Conference on Rehabilitation Robotics June 24-26, 2013 Seattle, Washington USA 978-1-4673-6024-1/13/$31.00 ©2013 IEEE