Life-Long Infrastructure Free Robot Navigation

In the future, autonomous vehicles will play an important part in our lives. They will come in a variety of shapes and sizes and undertake a diverse set of tasks on our behalf. We want smart vehicles to carry, transport, labour for and defend us. We want them to be flexible, reliable and safe. Already robots carry goods around factories and manage our ports, but these are constrained, controlled and highly managed workspaces. Here the navigation task is made simple by installing reflective beacons or guide wires. This project is about extending the reach of robot navigation to truly vast scales without the need for such expensive, awkward and inconvenient modification of the environment. It is about enabling machines to operate for, with and beside us in the multitude of spaces we inhabit, live and work. Even when GPS is available, it does not offer the accuracy required for robots to make decisions about how and when to move safely. Even if it did, it would say nothing about what is around the robot and that has a massive impact on autonomous decision-making.Perhaps the ultimate application is civilian transport systems. We are not condemned to a future of congestion and accidents. We will eventually have cars that can drive themselves, interacting safely with other road users and using roads efficiently, thus freeing up our precious time. But to do this the machines need life-long infrastructure-free navigation, and that is the focus of this work.We will use the mathematics of probability and estimation to allow computers in robots to interpret data from sensors like cameras, radars and lasers, aerial photos and on-the-fly internet queries. We will use machine learning techniques to build and calibrate mathematical models which can explain the robot's view of the world in terms of prior experience (training), prior knowledge (aerial images, road plans and semantics) and automatically generated Web queries. The goal is to produce technology which allows robots always to know precisely where they are and what is around them. Robots have a big role to play in our future economy, but underpinning this role will be life-long infrastructure-free navigation.

Impact Summary It is hard to understate the importance of the transport of goods and people in daily life. We depend on it totally. Any increase in efficiency, access, safety or reliability will have a major economic and societal impact. This proposal aims to achieve just those things. We will use information engineering, computing and robotics to provide a low cost underpinning for smart vehicles in civil, defence and industrial domains. Such vehicles offer the possibilty of end-to-end goods transportation - from raw materials to point of sale. They promise improved efficiency and safety on our roads. They give our aged, infirm and sensorially impaired citizens the hope of independent personal transport. Our aim is to enable this without requiring new navigation infrastructure. Car manufacturers have long been interested in improving driver experience and safety through sensing and processing. The vision of this proposal reaches beyond better parking sensors or lane detectors to real automotive autonomy and assisted driving. Robotics science has a massive role to play. In 2008, 2000 people were killed on UK roads due to driver error or concentration loss. We can address this by building cars which interpret their surroundings hundreds of times a second, never grow weary and access the cumulative experience of all cars. We need not wait for complete autonomy. The path to driverless transport is rich with exploitation opportunities. For example, smart cars can aid congestion control systems where humans currently respond to flow control signals in overly cautious and far from optimal ways. Cars that control themselves efficiently will improve flow, reduce congestion, pollution and transit times. There are energy considerations too. The CO2 cost of manufacture of the average sedan is 11% of its total life footprint. A fleet of autonomous cars operating ceaselessly in our cities would increase transport capacity and increase the person miles/kg CO2. Access to independent transport is another huge motivation. We insist, with good reason, that only the able can drive cars. We often preclude the infirm, partially-sighted and otherwise disabled, demanding that their access to roads be dependent on the munificence of others and public transport. Ultimately, autonomous cars, capable of operating on our roads as they exist now, operating beside and amongst traditional cars will extend the reach of transport to these disadvantaged citizens. This research is driven by this ultimate goal and the belief that this can be achieved by putting smarts in individual cars alone rather than on the roads themselves. The UK Department for Transport is a partner in this project and will ensure relevance to the UK's transport and logistics portfolio. Industry already benefits from precision robot navigation. For years now factories have been serviced by automated guided vehicles (AGVs) but they require awkward installation of beacons or buried wires. Using similar methods, ports in Australia and Singapore are fully automated. Infrastructure-free navigation will benefit AGV producers through market generation and users through increased efficiency. We will see increased coverage, cheaper operation and greater flexibility. There is a prima-facie need for autonomous vehicles in the defence sector. Our armed forces are keen to use smart robots for urban reconnaisance in anti-terrorist and rescue roles. The importance of mobile autonomy in national security is made clear by a US Congressional Mandate that an astounding one-third of all ground vehicles in the US Armed Forces will be unmanned by 2015. Impact will be driven by the involvement of partners representing automotive (Nissan), defence (BAe), industrial (Guidance Ltd), sensing (Navtech), international academic (MIT) and policy-making (DfT) sectors. It will be managed through licensing and inter-partner coorperation. It will be generated by first-class robotics research.