Biography
Julie joined the ORI in July 2014 as a DPhil student and was a member of Pembroke College. Her main area of study was Learning for Planning and Navigation.
Her work focused on context-based tracking to understand the dynamics of a scene given the environment the robot is in, and thus aid the planner in producing feasible and optimal paths through the environment. The world that surrounds us is in essence dynamic, and robots which operate in large-scale environments need to have the capacity to recognise high-level properties of the world, beyond the low-level and limited information retrieved from their sensors. This involves the ability to continuously learn from experiences, just as we humans effortlessly do.
Her interests lay in assistive robotics, where machine learning and medicine combine to aid patients with disability (smart exoskeletons and prosthetics, interfacing with the brain for stroke rehabilitation).
Julie holds a Master’s degree in Aerospace Engineering from ISAE-SupAero (Toulouse, France), and has previously worked at NASA Ames Research Center (California, USA) studying the climate of Mars.
In her free time, she enjoys playing the piano, engaging in team sports, and exploring the air, the land and the seas. She cherishes the dream of one day going to Mars with a pet robot.
Publications
- J. Dequaire, P. Ondruska, D. Rao, D. Zeng Wang, and I. Posner, “Deep tracking in the wild: End-to-end tracking using recurrent neural networks“, the International Journal of Robotics Research, 2017
- B. Yeomans, H. Porav, M.Gadd, D.Barnes, J.Dequaire, T. Wilcox, S. Kyberd, S. Venn, and P. Newman, “MURFI 2016 – From Cars to Mars: Applying Autonomous Vehicle Navigation Methods To a Space Rover Mission“, Advanced Space Technologies in Robotics and Automation, 2017
- J. Dequaire, D. Rao, P.Ondruska, D. Zeng Wang, and I. Posner, “Deep Tracking on the Move: Learning to Track the World from a Moving Vehicle using Recurrent Neural Networks“, ArXiv e-prints, 2016
- P. Ondruska, J. Dequaire, D. Zeng Wan, and I. Posner, “End-to-End Tracking and Semantic Segmentation Using Recurrent Neural Networks“, Robotics: Science and Systems Workshop, 2016 . Best Paper Award
- J. Dequaire, C. Hay Tong, W. Churchill, and I. Posner, “Predicting Localisation Performance in Teach and Repeat“,International Conference on Robotics and Automation, 2016
