Project/scholarship details

This interdisciplinary project, aims to develop theoretical and experimental frameworks for the design, control and evaluation of a new generation of networked assistive mobile robots (AMRs). Research in this area of human-centered robotics has been intense in the last years worldwide, as well as in our group at ISR-UC with state of the art results. However, despite all this effort, no commercially available intelligent AMRs, such as robotic wheelchairs, are currently available on the market. With the advent of novel and powerful sensors, with the development of novel path planning and machine learning algorithms, and due to the increasing capabilities of modern computers, some earlier bottlenecks for AMRs are likely to be eliminated. This project addresses human-centered mobile robotics and will contribute for safer and more user-friendly mobile robotic assistants adapted to human environments. In partnership with the APCC (Associação de Paralisia Cerebral de Coimbra), we aim to contribute with results towards better mobility of people suffering from neuromotor disorders. Human factors will be taken into account, namely by benchmarking and evaluating the developed approaches by end users. Human-machine interfaces are a central key in Human-centered mobile robotics since they define how users can input commands to steer the mobile robot. Users with severe motor disabilities need interfaces that can be controlled with minimal or zero muscular activity, such as brain-computer interfaces and eye-trackers. However, this type of interface provides information that is sparse in time and that may be unreliable. The use of multi-modal interfaces can help to increase both information transfer rate and reliability, but it is still not enough to operate efficiently a mobile robot (e.g., a wheelchair) in domestic environments. Therefore, Human-machine shared-control navigation, accepting input commands from user is required to have a safer and efficient navigation. The development of 2D/3D perception/reconstruction is a major requirement for SLAM and to plan navigation trajectories in cluttered and dynamic environments.

  • Funder

    FCT - Fundação para a Ciência e a Tecnologia, I.P.

  • Funder's country

    Portugal

  • Funding program

    5876-PPCDTI

  • Funding amount

    437,558.00 €

  • Start date

    2013-01-01

  • End date

    2015-12-31

Documents

go first page go previous page
go to page: 1
go next page go last page
rss feed search results

Learning Under Distribution Mismatch Applied in Biosignal Processing

Khalighi, Sirvan

Tese de Doutoramento em Engenharia Electrotécnica e Computadores, na especialidade de Automação e Robótica, apresentada à Faculdade de Ciências e Tecnologia da Universidade de Coimbra; Bioelectrical signals, which records brain activity, are among the complex dynamic signals due to the strong non-stationarity effects of brain and subject dependency. In biosignal-based classification, training samples and unlabe...

Date: 2016   |   Origin: Estudo Geral - Universidade de Coimbra
More info.

Dynamic Bayesian network for semantic place classification in mobile robotics

Premebida, Cristiano; Faria, Diego R.; Nunes, Urbano

In this paper, the problem of semantic place categorization in mobile robotics is addressed by considering a time-based probabilistic approach called dynamic Bayesian mixture model (DBMM), which is an improved variation of the dynamic Bayesian network. More specifically, multi-class semantic classification is performed by a DBMM composed of a mixture of heterogeneous base classifiers, using geometrical features...

Date: 2016   |   Origin: Estudo Geral - Universidade de Coimbra
More info.

Calibration, Structure-from-Motion and Registration Beyond Point Features

Raposo, Carolina dos Santos

Tese de doutoramento em Engenharia Electrotécnica e de Computadores, na especialidade de Automação e Robótica, apresentada ao Departamento de Engenharia Electrotécnica e de Computadores da Faculdade de Ciências e Tecnologia da Universidade de Coimbra; Geometric computer vision is strongly based in point primitives in problems of calibration, Structure-from-Motion (SfM), and registration. The reason for this is ...

Date: 2017   |   Origin: Estudo Geral - Universidade de Coimbra
More info.
go first page go previous page
go to page: 1
go next page go last page
rss feed search results

Report

Create a report with all publications from the project or scholarship in the current date for FCT scientific report.


Embed

Dynamically incorporate publications in your site (HTML) 

								
<script type="text/javascript">
	<!--
	document.write('<div id="rcaap-widget"></div>');

	document.write('<script type="text/javascript" src="https://www.rcaap.pt/snippet?resource=documents&project=FCT%2F5876-PPCDTI%2F126287&fields=id,titles,creators,issueDate,link,descriptions"></script>');
	-->
</script>