Development & interpretation of a perception sensory system for real-time robotic aided therapy of autistic children

• Grant number: 13/26025-0
• Support Opportunities: Scholarships abroad - Research Internship - Master's degree
• Start date: March 20, 2014
• End date: September 19, 2014
• Field of knowledge: Engineering - Biomedical Engineering - Medical Engineering
• Principal Investigator: Fernando Santos Osório
• Grantee: Leonardo Milhomem Franco Christino
• Supervisor: Kaddour Bouazza-Marouf
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: Loughborough University, England
• Associated to the scholarship: 13/13880-9 - GPU service: GPU acceleration of robotic systems services focused in real-time processing of 3D point clouds, BP.MS

Abstract

In typical developing infants, cognitive and emotional attitudes towards objects and people constitute the ingredients of social relationships. Infantile autism, which is a neurodevelopmental disorder, is characterized by difficulties in social and interpersonal communication as well as in processing own and other people's emotions. In autism rehabilitation therapy, different approaches are currently being utilized to better understand the capacity of autistic children for social and emotional interaction. These approaches are based on the belief that artificial environments i.e., robots, seem to be more helpful than real environments in allowing autistic children to express social interest; this in turn could lead to development of their social skills. Currently, the robot which is used to social-interact with these children is being fully teleoperated by the therapist depending on the child's behavior. The aim of the project is to develop the perception and reaction of the robot using image and depth sensors, making the robot able to make some decisions automatically, such as obstacle avoidance or reaction on the child's action, as well as proposing new automatic interactions depending on the child's gestures. These characteristics add a lot of value to the already-in progress project being executed. The perception will be accomplished using 3D point cloud based sensors (e.g. Kinect), and implementing modules (as services) that will allows us to segment and tracking the main scene elements (e.g. obstacles, child) in real-time. (AU)