Last week, two of our academics presented the cutting-edge research performed in LCAS at IROS, one of the major robotics conferences in the world, hosted this year in Kyoto.

In particular, Francesco Del Duchetto presented “Learning on the Job: Long-Term Behavioural Adaptation in Human-Robot Interactions“, which was also selected as a finalist for the best paper award in Cognitive Robotics. In this work, Francesco introduces a reinforcement-based framework that allows a robot to modify and adapt its own behaviour to accommodate and improve human interactions.

At the end of the conference, as part of the “Perception and Navigation for Autonomous Robotics in Unstructured and Dynamic Environments” (PNARUDE) Workshop, Dr Riccardo Polvara presented “Collection and Evaluation of a Long-Term 4D Agri-Robotic Dataset“, an ongoing research underpinned within the EU H2020 BACCHUS project to further improve long-term autonomous operation of mobile robots in the agricultural domain. In this work, Riccardo describes a novel method for pruning a pointcloud out of those points which are not stable through the years (e.g., those associated with the foilage) so to lead to improved localization performance and a reduction of mapping cost. 

Few pictures from the conference: