• assistive techonolgy
• disability
• embedded devices
• artificial inteligence
• robotics
• biomedics

The present PhD Thesis work adresses the development of electronics technologies for people with disabilities. After brefly introducing the concept of assistive technology and who uses it, we present some of the most important socio-economic aspects of the assistive technology scenario and why R\&D in universities and public research centers are so important in this field, more than in others.

The main objectives of the work is to show the current researches in Assistive Technology field carried out by the candidate. In the following chapters these research topics will be discussed:


- In chapter 2 we explore the design of audio algorithms, hardware IP and devices for people with hearing losses. Our systems exploit microphone arrays to enhance acquired audio by using advanced state of the art audio algorithms like beamforming and noise reduction. Also a lot of effort was devoted to build these systems with minimal power consumption.
- In chapter 3 it is shown the design of a wheelchair mounted robotic arm addressed for people with motor skill impairments, the project RIMEDIO. We build a fully autonomous robotic arm that can perform simple tasks like knocking a door, press an elevator button, switch on/off the light and similar tasks.
- In chapter 4 we explore a state of the art algorithm for 3D scene reconstruction. The proposed scene reconstruction pipeline is capable of automatically reconstruct full 3D views of the environment even with a single viewpoint by inferring similarities in geometries and shapes between known objects. This system can be a base building block for fully autonomous robotic manipulators.
- In chapter 5 it is described a wearable intelligent computer vision system intended for visually impaired users. The system is designed to recognize the objects within the images and it currently supports TensorFlow, a deep learning software framework released by Google, and it is especially designed for pattern recognition activities.
- Finally, in chapter 6 we present a wearable devices specifically tailored for people with neurogenic bladder. Patients with impaired bladder volume sensation have the necessity to monitor bladder level in order to avoid urinary tract infections and urinary reflux that can lead to renal failure. Our wearable solution uses bioimpedance measurement to monitor bladder volume level. Adaptive intelligent algorithms can improve performance and accuracy of our system.


Each chapter introduces a real world problem that was brought to our attention by the users. We apply state of the art technologies and methodologies from engineering and computer science to solve it. Stakeholders as clinicians, research centers, devices manufacturers, IT multinationals and final users were involved at different stages in the development process.

All the research activities presented in this work are responding to a new demand for AT devices by people with disabilities and try to give solutions to the new challenges on Assistive Technologies.