Abstract
The Safe Home Care project focuses on assembling safe home assisted-living environments built on autonomous Off-The-Shelf systems. We argue that these smart spaces will contribute to relieve the pressure on health systems by providing the means for ambulatory and daily life assistance. However, the integration of disparate sovereign systems will not be easy to accomplish since the number of interaction scenarios will be impossible to predict and evaluate a priori. Therefore, we propose the SHC reflective middleware framework, conceived with two goals in mind: i) manage the safe integration of off-the-shelf systems (cf. interference-free) by exploiting reflection and 3D virtual world simulation; ii) provide non-intrusive pervasive interface mechanisms for home assisted-living actors. In this paper, we focus specifically on the first goal by providing the means for generating 3D simulations; the states generated during simulation are then analyzed by a graph-pruning algorithm to perceive feature interactions in pre-deployment phases. We evaluate our approach on specific home care use cases.

Abstract

Abstract

Abstract
This paper presents an interior design application that uses photos from real indoor spaces and rooms as input. The user is encouraged to take a picture of a certain place and virtually reshape it with furniture, different colors, textures and materials. The system analyses the still image of the physical space through image processing and computer vision algorithms to detect the world orientation relative to the camera. Knowing this, it is possible to create a natural interface where objects are laid on the floor and pushed around as in real life.

Abstract
This paper proposes two extensions of Particle Swarm Optimization (PSO) and Darwinian Particle Swarm Optimization (DPSO), respectively named as RPSO (Robotic PSO) and RDPSO (Robotic DPSO), so as to adapt these promising biological-inspired techniques to the domain of multi-robot systems, by taking into account obstacle avoidance. These novel algorithms are demonstrated for groups of simulated robots performing a distributed exploration task. The concepts of social exclusion and social inclusion are used in the RDPSO algorithm as a punish-reward mechanism enhancing the ability to escape from local optima. Experimental results obtained in a simulated environment show that biological and sociological inspiration can be useful to meet the challenges of robotic applications that can be described as optimization problems (e.g. search and rescue). © 2011 IEEE.

Abstract
Fractional calculus (FC) is being used in several distinct areas of science and engineering, being recognized its ability to yield a superior modelling and control in many dynamical systems. This article illustrates the application of FC in the area of robot control. A Fractional Order PD mu controller is proposed for the control of an hexapod robot with 3 dof legs. It is demonstrated the superior performance of the system by using the FC concepts.