GNB2014, June 25th-27th 2014, Pavia, Italy 1 Abstract— This paper investigated the opportunity to outsource demanding services in the cloud for assistive robotics and Ambient Assisted Living. A cloud robotics solution was designed and preliminary tested to manage and pull together a mobile robotic platform and a smart environment, in order to provide “personal safety and healthcare management at home ” service. The solution was evaluated in terms of Quality of Service (QoS) and tested in the realistic scenario of the DomoCasa Living Lab, Peccioli, Italy. A localization algorithm was outsourced in the cloud and pulled data from a smart environment. Experiments demonstrated a service’s Reliability of Respond at least of the 0.04% and a Time of Respond of the same order of magnitude of the processing time required by the user localization algorithm. The robotic reminding service was provided in less than 22 seconds. Keywords—Cloud robotics, AAL Service Design, Service Robot, Quality of service. I. INTRODUCTION n the field of the Ambient Assisted Living (AAL), many projects have focused on the development of service robots able to support senior citizens in their daily activities [1]. As a result, many standalone and networked robots, also integrated in smart environments, have been developed to perform specific tasks [2-3], act as simple companion robot [4-5] or provide complex assistive services [6-7]. At the Humanoids 2010 Conference, James Kuffner, professor at Canergie Mellon University, defined Cloud Robotics as “the combination of cloud computing and robotics”. In his paradigm, cloud robotics is not related to a specific type of robot, but to the way robots store information and access a base knowledge. As a matter of fact, cloud computing could give to robotic systems the opportunity to exploit user centered interfaces, computational capabilities, on-demand provisioning services and large data storage with minimum guaranteed QoS, scalability and flexibility. In authors’ opinion, Cloud Service Robotics (CSR) could be defined as the integration of different agents that allow an efficient, effective and, robust cooperation between robots, smart environments and humans, to provide continuous services to senior citizens in their daily living environments in a scalable, affordable and reliable manner. Cloud robotics paradigm is nowadays applied in many robotic applications, enabling robots to offload CPU-heavy tasks and access base knowledge to expand robot consciousness beyond their physical body [9-10]. In literature [11-13], Localization Based Services (LBSs) were considered resource demanding and several works have investigated the advantages of the cloud for a QoS improvement. In standalone robot, the user position could be computed by means of high cost sensor embedded on the robot that use complex algorithms to discover and reach users; in networked robots, wireless sensor networks installed in the environments or worn by users could be used to share information about localization and facilitate the robot to discover and reach the users. Instead in a cloud robotic solution, the system could take advantage of running complex algorithms on the cloud platform that could act as a modular and always connected remote brain that improved the Quality of Services (QoS) [8]. In this context, the aim of this paper is to present a cloud robotic solution that allow to enhance the QoS provided by complex systems composed of smart environments and robotic devices. The solution focused on the evaluation of the QoS on a specific service, named “personal safety and healthcare management at home“. II. METHODS AND INSTRUMENTS The proposed solution integrated a mobile robotic platform and a smart environment by means of a cloud platform. The system provided the following services to the user:  critical situations recognition at home, by means of environmental sensors that alert locally the old person and/or remotely relatives or professional caregiver that can immediately control status of the home using a smartphone or tablet;  remind the old person about pharmacological therapies or medical treatment schedule, such as visit at hospital, by means of a robotic system. A Software as a Service (SaaS) cloud infrastructure was designed as described in Fig 1. The infrastructure was composed of four functional layers: (1) Hardware (2) Communication, (3) Application and (4) Interface layers, to provide required services to the users. Hardware layer consisted of a mobile robotic platform and 2 ZigBee-based Wireless Sensor Networks (WSNs), one for user localization (Localization Network - LNet), and the other for environmental monitoring (Sensor Network - SNet). The robot was based on a SCITOS G5 platform (Metralabs, Germany) that communicated with the user by means of an embedded touch screen, speech recognition and voice synthesis module and colored LED lights for visual feedbacks. The robot received from the cloud the proper instructions to accomplish the required services and data about the user position, using an embedded Wi-Fi module. The WSNs were developed for home monitoring and user localization and were composed of a network coordinator, a data logger node and a set of Sensor Nodes (SN). Each SN contained a selection of sensors described in Fig 2. Communication layer was composed of two complementary tiers [10]: Machine-to-Machine (M2M) and Machine-to-Cloud (M2C) communication modules. ZigBee, WI-FI and TCP/IP protocols were used for M2M communication to directly exchange data between the agents Design of a Cloud Robotics Solution for Ambient Assisted Living Application L. Fiorini 1 , M.Bonaccorsi 1 , F. Cavallo 1 , R. Esposito 1 , and P. Dario 1 1 The BioRobotics Institute, Scuola Superiore Sant’Anna I