New Oxford Robotics Institute website coming soon!
The Oxford Robotics Institute is the next step in the evolution of the Mobile Robotics Group.
Mobile Autonomy: A Pervasive Technology
The Oxford Mobile Robotics Group researches all aspects of land based mobile autonomy. We invent the technologies that allow machines to ask and answer “Where am I?”, “What surrounds me?” and “What should I do next?” These three key questions underpin all that we do. They force us to confront fundamental questions in navigation, perception, machine learning and systems design. We are proud of the way we validate our thinking and challenge ourselves by deploying and running large robotics systems. These “flagship” projects, drawn from many application domains are central to our mission – we solve real world robotics problems. In 2014 a spin-out company called Oxbotica was launched to drive commercialisation and cross domain impact of our work.
ORI wins best student paper award at IROS
Congratulations to Markus Wulfmeier, Dominic Wang and Ingmar Posner on winning the best student paper award for their work on Deep IRL: Watch This: Scalable Cost-Function Learning for Path Planning in Urban Environments – an approach to learning cost maps for driving in complex urban environments. The authors used an ambitious dataset collected over the course of a year of driving in pedestrianised areas in Milton Keynes.
Self-driving vehicles tested in public
Self-driving vehicles equipped with Oxford-developed autonomy software have been tested successfully in public for the first time in the UK. The demonstration in Milton Keynes was coordinated by the Transport Systems Catapult (TSC) and marked the conclusion of the LUTZ Pathfinder project, which has been running for the past 18 months.
Our Q14 Rover goes to Space (nearly)
ORI is collaborating with thirteen other institutions under the auspices of The UK Space Agency and in partnership with the Canadian Space Agency in a major Mars Exploration Analogue mission taking place over several weeks in the Utah desert this autumn. The mission will be coordinated from Harwell and will comprise exploration and planetary science activities undertaken from ORI’s Q14 Rover. We are providing the Rover together with engineering support and in addition to Rover operations will have the opportunity to implement and test our navigation algorithms under realistic Mars like conditions. Currently the Rover is being re-engineered to prepare it for duty and to extend the operator software interface.
ORI is recruiting
We are looking for academics and researchers to join us – more details here.
MRG at the Shell Eco Marathon
We took our Shell vehicle to the London Eco Marathon and triumphed with the first autonomous laps on a Shell Eco Marathon track and the first non-MRG safety driver – Kimi Räikkönen – who achieved two autonomous laps after training in Selenium autonomous technology.
Watch Principal Investigator Ingmar Posner being interviewed by Car Throttle Extra here.
MRG awarded Best Paper in the RSS 2016 Deep Learning Workshop
Congratulations to Peter Ondruska, Julie Dequaire, Dominic Zeng Wang, and Ingmar Posner for their Best Paper award in the RSS 2016 Workshop: Are the Sceptics Right? Limits and Potentials of Deep Learning in Robotics. Their paper “End-to-End Tracking and Semantic Segmentation Using Recurrent Neural Networks” presents a novel end-to-end framework for tracking and classifying a robot’s surroundings in complex, dynamic and only partially observable real-world environments.
Read more here.
JLR Range Rover delivered to MRG
JLR have kindly loaned a Range Rover to MRG for use in autonomous driving development. We are now kitting it out with cameras and sensors and will start trials shortly.
UK sets the pace on autonomous vehicles
If you head down to the O2 in Greenwich later this year to catch an event, you may notice an unusual attraction winding its way through the crowds of tourists. Seven fully electric, fully automated shuttles will be carrying people around the peninsula. They will navigate autonomously, using a combination of lasers, sensors, cameras and some very clever software developed by UK researchers …
Philip Rutnam, Perm Secretary for the Department for Transport and Civil Service Disability Champion writes about our autonomous navigation technology in his blog. You can read more here.
LUTZ Pod Update
We are looking forward to delivery of the first LUTZ Pod. After initial delivery to us in late 2015, it went back to RDM in Jan 2016 for some design, software and manufacturing fixes. Delivery of a working base vehicle will be a great milestone in this project, which is a collaboration between MRG, RDM, who make the base hardware, and the Transport Catapult.
When the vehicle is delivered we will be installing an autonomy system in it. This involves installing a whole suite of sensors around the vehicle – we use cameras, lasers and radars. Just as in the RobotCar project, these are connected to a set of processing modules within the vehicle that run the autonomy software. This combination of sensors and software allows the vehicle to learn the structure (map) of its environment and then when operating navigate, plan and operate within it to carry passengers. It does this while watching out for and coping with hazards like cars, people, bikes and kerbs. Watch this space for information on the LUTZ pods.
Best Student Paper Award Finalist – ICRA 2016
Congratulations to Michael Tanner, Pedro Piniés, Lina María Paz, and Paul Newman on their selection as one of three finalists (of 817 total papers) for the Best Student Paper award at the 2016 International Conference on Robotics and Automation (ICRA). Their paper “What Lies Behind: Recovering Hidden Shape in Dense Mapping” presents a method to reconstruct a city’s underlying 3D architecture using only a single push-broom LIDAR traversal.
Ingmar Posner – Outstanding Supervisor
We are delighted to announce that Professor Ingmar Posner has just won the OUSU Teaching Award 2016 for Outstanding Supervisor in the Mathematical, Physical and Life Sciences Division.
Ingmar with DPhil students Dan, Hillary, Corina, Martin, Julie and Markus.
UK Robotics Week June 2016
This year’s UK Robotics Week takes place from 25 June – 1 July 2016, with kick-off at Imperial College London 7-8 May. Read more details here.
Upcoming Oxford RobotCar Dataset announced
Setting a new benchmark in real-world data, the Oxford RobotCar dataset contains over 100 repetitions of a consistent route through Oxford, UK, captured over a period of more than a year. The dataset captures many different combinations of weather, traffic and pedestrians, along with longer term changes such as construction and roadworks. We plan to make it available for download in the second half of 2016.
More information here.
Mobile Robotics Week – CDT Challenge
11 of Oxford’s AIMS (Autonomous Intelligent Machines and Systems) CDT students joined MRG to take part in a robotics challenge, working with 2nd and 3rd year DPhil students to programme a self-driving robot.
In the final challenge each robot had to work its way around obstacles to reach a goal.
Points were awarded for speed and accuracy and deducted for hitting the obstacles.
All teams worked well – but these were the winners.
For more information on the AIMS Centre for Doctoral Training see here.
Jo Johnson visits MRG
Jo Johnson with Professors Paul Newman and Ingmar Posner
Universities and Science Minister Jo Johnson visited MRG on 29th February to see world-leading research and announce new funding.
Jo Johnson with Professor Ingmar Posner and DPhil student Julie Dequaire
The Universities and Science Minister, Jo Johnson, visited the University’s Networked Quantum Information Technologies (NQIT) Hub and Mobile Robotics Group on 29th February. His tour of the MRG workshop included demonstrations of MRG’s latest technology and meetings with DPhil students Julie Dequaire, Dan Barnes and Markus Wulfmeier.
DPhil student Dan Barnes demonstrates MRG’s ground-breaking technology
During his visit to Oxford he announced new funding of over £200 million to support PhD students in engineering and physical sciences and to boost the UK’s research into quantum technologies, saying:
“We are committed to securing the UK’s position as a world leader in science and innovation… This new funding builds on our protection for science spending by supporting research in our world-leading universities and helping to train the science leaders of tomorrow.”
Julie Dequaire, a second year PhD student at MRG commented: “It is an incredible opportunity for me to be here … it is great to see how dynamic robotics is in the UK. We have robotics that can easily adapt in the real world and that is something we are very lucky to have. This is why we need this kind of funding to keep that up.”
OU Professor of Information Engineering Paul Newman said: “This kind of money is vital for the nation in the next decade in terms of developing in areas of science and engineering. It’s a great show of the Government to do this, it’s all part of backing the nation to be technology leaders.”
MRG’s workshop – with DPhil student Markus Wulfmeier and Senior Researcher Chi Hay Tong
Digital Railways project update
We have just completed the latest InnovateUK project review on our Digital Railways project, which sees our technology deployed on trains in order to provide robust and scalable detection and mapping of railway assets. This project brings together a large part of MRG technology from mapping and localisation at high speeds to calibration, active learning and object detection. After a number of successful technology trials on the railways the project is rapidly entering its penultimate quarter and we look forward to gathering some high speed data with our partners.
Best Paper Award – ICCV 2015
Congratulations to Geoff Pascoe, Will Maddern and Paul Newman on winning the Best Paper Award for their paper Direct visual location and calibration for road vehicles in changing city environments at 3rd Workshop on Computer Vision for Road Scene Understanding and Autonomous Driving, ICCV 2015.
Self-driving pods win national innovation award
The LUTZ Pathfinder project saw off competition from four other shortlisted contenders to pick up the prestigious trophy at the 99th SMMT Annual Dinner.
Overseen by the Transport Systems Catapult on behalf of the UK Automotive Council, LUTZ Pathfinder is a collaborative project involving Coventry-based manufacturers RDM Group, the University of Oxford’s Department of Engineering Science Mobile Robotics Group (MRG) and Milton Keynes Council.
The three electric-powered pods will be used not only to trial the technology involved in operating self-driving vehicles, but also to assess public response to the pods in a real-world setting.
The first LUTZ Pathfinder pod is currently at Oxford MRG for the integration of the autonomous control system.
Read the full article here: