Where Vehicles Think For Themselves: Oxbotica

Oxbotica, a leading autonomous software company based in Oxford, is bringing self-driving technology to the roads, ports, mines, roads, forests, quarries of the world (and even to Mars!). Its autonomous operating system, Selenium, is platform agnostic and navigates autonomously around cities, warehouses and off-road environments, using data from lasers and cameras placed on vehicles. Originally founded in 2014 as a spin-out of Oxford University’s Oxford Robotics Institute, Oxbotica helps its clients integrate over 200 man-years’ worth of work in autonomy into their own products.

The company’s solutions allow robots, vehicles, machinery and people to precisely map, navigate and actively interact with their surroundings, delivering new capability and precision to a wide range of applications. Oxbotica’s 3D imaging and localisation solutions operate indoors and outdoors and are suitable for use in applications ranging from handheld survey devices to autonomous vehicles.

Sneak Peek at the Projects

Oxbotica is leading a consortium of companies that will help to further cement the UK’s reputation as a world leader in the development of autonomous vehicles.

DRIVEN: DRIVEN is a £12.6 million project – all about insuring, ensuring and exporting fleet wide Level 4 connected autonomy. Bringing together dedicated autonomy specialists, world-class innovators, key enablers, disseminators, and transport experts, DRIVEN is the most complex CAV trial that’s ever been attempted.

CargoPod: CargoPod is an autonomous last-mile delivery vehicle, developed by Oxbotica. The van is guided by its state-of-the-art autonomy software system Selenium, which enables real-time, accurate navigation, planning and perception in dynamic environments.

The Geni: Geni is Oxbotica’s prototype development vehicle. It is one of the main platforms on which the team tests out new software. It is an electric vehicle that is based on Oxbotica’s main workshop and test track in South Oxfordshire. It runs Oxbotica’s autonomous operating system – Selenium.

Lutz Pathfinder: Self-driving vehicles equipped with Oxford-developed autonomy software were tested successfully in public for the first time in the UK in October. The demonstration in Milton Keynes was coordinated by the Transport Systems Catapult (TSC) and marked the conclusion of the LUTZ Pathfinder project, which ran for 18 months.

GATEway: Oxbotica provided both the autonomy system, Selenium, and the fleet management system, Caesium, to the initial proof of concept prototype vehicle in the GATEway project (Greenwich Automated Transport Environment). This is a £8m research project, led by TRL and jointly funded by Innovate UK and industry, to understand and overcome the technical, legal and societal challenges of implementing automated vehicles in an urban environment.

A Board Gauge Look at Oxbotica’s Products

Selenium: Selenium is Oxbotica’s autonomous control system, a vehicle agnostic operating system that can work on anything from forklifts, to cargo pods, to vehicles. The software uses the knowledge of where it is in the world, together with local information about the environment around the vehicle, to determine a safe path and velocity to move the vehicle towards its goal. Oxbotica is providing its Selenium autonomous control system to the GATEway project in Greenwich, London.

NABU: NABU, Oxbotica’s Navigation Base Unit, is a 3D imaging device which uses patented laser and vision integration to rapidly scan an indoor or outdoor environment. This is both a disruptive technology and a cost-reduced technology – enabling new applications as well as enabling access to previously unaffordable technology for many survey and localisation companies. Many of the core localisation, mapping and perception algorithms integrated into Selenium can be found running on NABU.

Warehousing: Oxbotica’s autonomous operating system, Selenium, can work in both outdoor and indoor environments. As Selenium is not reliant on GPS to operate, Selenium can be used in warehouse environments for mapping, asset management, and 3D modelling. When integrated onto autonomous forklifts, just by using camera data, our system can understand where it is in the world, what’s around it, and work out where to go next.

The company’s vision-only localisation system, Dub4, is extremely robust to changes in appearance and lighting, and unlike many other warehouse mapping and localisation systems, copes well with structural changes in the environment. Whether these changes are moving shelves, people or other forklifts, Selenium can figure out where it is and where to go next.

Caesium: Caesium is Oxbotica’s fleet management system, a cloud-based service that schedules and co-ordinates fleets of autonomous vehicles. It enables route optimisation and data exchange between the vehicles without human intervention.

Caesium is used on the GATEway shuttles, showing shuttle operator information about the shuttle, such as what mode it’s in (manual or autonomous), its velocity and heading, the amount of battery remaining, and the temperature of key components. It also shows its next destination. From the iPad that’s running Caesium, an operator can cancel autonomy at the touch of a button, or request maintenance.

The Thought leaders

Graeme Smith, CEO: Graeme is Oxbotica’s CEO, with a substantial track record in delivering complex products and services from research and development through to customer launch – primarily in the automotive sector, working at Ford and Connexis.

Paul Newman, Founder: Paul founded Oxbotica, and is also Principal Investigator at the University of Oxford’s Oxford Robotics Institute. His focus lies in pushing the boundaries of navigation and autonomy techniques in both endurance and scale.

Ingmar Posner, Founder: Ingmar founded Oxbotica, and is also an Associate Professor in Information Engineering at the University of Oxford. He leads the perception and planning work within the Oxford Robotics Institute at the University of Oxford.

“We’re developing the next generation of autonomous vehicles - creating the software that makes them go. Using the latest in computer vision and machine learning, our systems learn from their environment and share experiences with each other, so that they’re getting smarter all the time.”