Robotics researchers are paving the way for the first footsteps on Mars, and making advances that could help save lives on Earth.
Working with a robot built by NASA may not sound like an ordinary day at the office to most of us, but for staff in Edinburgh’s School of Informatics, it is becoming so.
Research teams led by Dr Maurice Fallon and Professor Sethu Vijayakumar are leading the field of robotics into uncharted territory through their work on Valkyrie, one of the most advanced humanoid robots in the world.
In March 2016, researchers began carrying out pioneering work on the state-of-the-art machine in an effort to help America’s space agency achieve its vision of sending Valkyries on missions to Mars.
Edinburgh is one of just three universities in the world, and the only centre in Europe, to have received a Valkyrie.
Constructed in 2015 and named after the female spirits of Norse mythology, the humanoid stands almost six feet tall and weighs 125kg.
With sophisticated on-board sensors and 44 moveable joints, Valkyrie can walk on two legs and carry out actions, such as picking up and manipulating objects.
“We’re pushing the limits of what can be achieved in the field of humanoid robotics,” says Dr Maurice Fallon, Chancellor’s Fellow in Robotics and Computer Vision.
“The research we are carrying out on Valkyrie presents unique and complex challenges, but with that comes the remarkable opportunity to explore a vast frontier in robotics. It’s valuable, exciting work.”
As part of future missions to Mars, NASA aims to use Valkyrie robots to set up and maintain equipment on the planet’s surface ahead of the deployment of human astronauts. Once humans arrive, the robots must be able to work closely and safely alongside them.
The Edinburgh team – which includes PhD students – will provide the humanoid with the detailed and sophisticated skills it requires.
“Sending Valkyries on pre-deployment missions makes space exploration safer and cheaper,” explains Dr Fallon.
“To make this possible it’s essential that humanoids are able to carry out all of the tasks and operations that a human astronaut would be expected to perform.”
Using their world-leading expertise in motion planning, machine learning and sensing, members of the team are developing software that will help improve Valkyrie’s ability to handle objects and manoeuvre on uneven terrain.
The team is also working on ways to overcome one of the big challenges of space exploration – time delay.
It would take around 30 minutes for signals from Earth to reach Mars, making functional human control of Valkyrie virtually impossible.
Progress we make on Valkyrie could aid in the development of exoskeletons for people with disabilities or serious spinal injuries. We’re also interested in enhancing the abilities of robots for use during disaster situations, such as following earthquakes and major industrial accidents.
Professor Sethu Vijayakumar, Chair of Robotics, explains: “We are seeking to develop Valkyrie’s own ability to make sense of the world around her and adapt to unforeseen events while still allowing high-level human input – a concept known as shared autonomy.
“Once on the surface of Mars, humanoids must be able to react in real-time to meet the challenges of operating in an ever-changing environment. For this to be achieved, we must provide Valkyrie with the tools she needs to think for herself.”
The humanoid is housed within the Edinburgh Centre for Robotics, a joint initiative with Heriot-Watt University, and is part of a long-term collaboration with NASA.
However, the overall aim of the research is much broader than sending a robot to Mars, and society could benefit from the team’s work on Valkyrie long before humanoids set foot on the Red Planet.
“We want to improve the core capabilities of robots to help improve, and in many cases save, people’s lives,” explains Professor Vijayakumar.
“Progress we make on Valkyrie could aid in the development of exoskeletons for people with disabilities or serious spinal injuries. We’re also interested in enhancing the abilities of robots for use during disaster situations, such as following earthquakes and major industrial accidents.”
The Valkyrie project is one of many worldleading research activities of the School, which is recognised as the highest quality informatics centre in the UK.
“Edinburgh has a reputation for excellence in informatics stretching back to the 1960s,” says Professor Johanna Moore, Chair of Artificial Intelligence and Head of the School.
Donald Michie, arguably most recognised for his contribution to code breaking at Bletchley Park during the second world war, set up the Department of Machine Intelligence and Perception at Edinburgh in 1963.
He developed one of the first programmes capable of learning to play a game, and invented the memoisation technique, a way of speeding up computer programmes.
In the same era, Sidney Michaelson was invited to set up a Computer Unit in Edinburgh, later being appointed to the first Chair of Computer Science in 1967.
Michaelson, working with KD Torcher, was one of the first computer scientists to develop microprogramming, and later with H Whitfield invented early versions of a multiuser operating system – commonplace today but unprecedented in the 1960s.
Professor Moore comments: “Some five decades later, the University of Edinburgh remains one of the world’s foremost centres for informatics research.”
The examples set by such pioneering and polymorphic computer scientists formed solid foundations for Edinburgh’s multidisciplinary approach to research and learning today.
For Professor Vijayakumar, the opportunities arising from this long-established recognition are second to none: “It’s testament to Edinburgh’s reputation as one of the world’s leading centres for informatics, that today we have the opportunity to work on Valkyrie and contribute to NASA’s goal of sending humanoids on missions to Mars in the future.”
Photo © Tricia Malley Ross Gillespie www.broaddaylightltd.co.uk