Come back home system for enhancing autonomy of individuals with acquired brain injury and dependent on their integration into society

This project has been funded by the Ministry of Economy and Competitiveness of the Government of Spain within the State Program of R+D+i Oriented to the Challenges of Society and co-financed with European Feder funds. The grant number is TIN2016-76515-R and its complete title: RETOGAR: RETORNO AL HOGAR: SISTEMA DE MEJORA DE LA AUTONOMÍA DE PERSONAS CON DAÑO CEREBRAL ADQUIRIDO Y DEPENDIENTES EN SU INTEGRACIÓN EN LA SOCIEDAD.

Abstract

In the last years, the care of dependent people, either by disease, accident, disability, or age, is one of the current priority research topics in developed countries. Moreover, such care is intended to be at patients home, in order to minimize the cost of therapies. Patients rehabilitation will be fulfilled when their integration in society is achieved, either in the family or in a work environment. To address this challenge, we propose the development and evaluation of an assistant for people with acquired brain injury or dependents. This assistant is twofold: in the patient’s home is based on the design and use of an intelligent environment with abilities to monitor and active learning, combined with an autonomous social robot for interactive assistance and stimulation. On the other hand, it is complemented with an outdoor assistant, to help patients under disorientation or complex situations. This involves the integration of several existing technologies and provides solutions to a variety of technological challenges. Deep leaning-based techniques are proposed as core technology to solve these problems.

Contribution

My specific contribution laid in the fulfilling of the "Outdoor patient assistant" task. This section focused on the development of an outdoor patient assistance system using wearable vision sensors and localization information provided by GPS to facilitate patient's navigation. Section 3.4.4 in the main paper.

The outdoor scene understanding pipeline was created using two deep learning systems that were jointly used: a semantic segmentation architecture and a monocular depth estimation model. First, a pixel-wise classifier performed inference on the wearable camera feed. This very same color frame and the previous one were both forwarded to a deep autoencoder that infers the correspondent depth map. The pixel-wise mask was then used to extract potential collision subjects such as bikes, pedestrians, or poles. The predicted depth was used to compute the distance to the obstacle.

Besides the creation of the assistance system also the novel outdoor datasets was contributed to conduct the task.