Skip to main content

Two papers nominated for awards at IEEE ICRA 2015

News Image

05 May 2015



More News...
ICRA is the IEEE Robotics and Automation Society's flagship conference where only 4 percent of papers out of about one thousand papers, received nominations.

ICRA awards page

The paper nominated for Best Cognitive Robotics Paper Award, Best Student Paper Award and Best Conference Paper Award is:

The Design and Evaluation of a Cooperative Handheld Robot
By Austin Gregg-Smith and Walterio Mayol

This paper is the first to describe about an exciting new type of robot: Cognitive Handheld Robots. That aim to be as intuitive as a handheld tool but with enhanced levels of motion, sensing and cognition. Essentially researchers at Bristol are trying to understand and develop a missing element in the "robot taxonomy tree". Robots that are neither independent (as is lets say a humanoid robot), nor wearable (as are exoskeletons). Handheld robots bridge the gap between these extrema and aim to cooperate with the user in order to solve tasks with the user providing some tactical guidance and the robot completing the low level detailed task.

An early video of the work is here.

Conscious that starting out a new area in robotics is hard, the team has made their designs open source and available from handheldrobotics.org

The second paper has been nominated for Best Robotic Vision Paper Award, Best Student Paper Award and Best Conference Paper Award and is:

Inverse Depth for Accurate Photometric and Geometric Error Minimisation in RGB-D Dense Visual Odometry by Daniel Gutiérrez-Gómez, Walterio Mayol, and Josechu Guerrero.

Both Daniel and Josechu are at the University of Zaragoza in Spain and this work is a result of the nice collaboration that started between Bristol CS and Zaragoza when Daniel made a research stay here last year.

In this paper we present a dense visual odometry system for RGB-D cameras performing both photometric and geometric error minimisation to estimate the camera motion between frames. Contrary to most works in the literature, we parametrise the geometric error by the inverse depth instead of the depth, which translates into a better fit of the distribution of the geometric error to the used robust cost functions. Importantly, we show that our approach is able to work in real time and we provide a qualitative evaluation on our own sequences showing a low drift in the 3D reconstructions.

ICRA 2015 will be held in Seattle, Washington, USA on 26th-30th May 2015.