Perceiving, Modelling and Interacting with the Object-Based World

Lead Research Organisation: Imperial College London
Department Name: Dept of Computing

Abstract

"Perceiving, Modelling and Interacting Autonomously in a Dynamic Object-Based World"

The Dyson Robotics Lab at Imperial College was founded in 2014 as a collaboration between Dyson Technology Ltd and Imperial College. It is the culmination of a thirteen-year partnership between Professor Andrew Davison and Dyson to bring his Simultaneous Localisation and Mapping (SLAM) algorithms out of the laboratory and into commercial robots, resulting in Dyson's 360 Eye vision-based vacuum cleaning robot in 2015 which can map its surroundings, localise and plan systematic cleaning pattern. Our success in working together made it clear that computer vision is a key enabling technology for future robots. This proposal aims to fund the Lab to push the forefront of visual scene understanding and vision-enabled robotic manipulation into new and more demanding application areas.

The research activity we are outlining in this Prosperity Partnership complements the large internal R&D investment that Dyson is making to to created advanced robotic products. The aims of this partnership are to invent and prototype the breakthrough robot vision algorithms which could truly take us to next generation capability for advanced robotics working in unstructured environments, and to transfer this technology into the long-term product pipeline of Dyson as they aim to open up new product categories.

Dyson has now been working on robotics for nearly 20 years, a period during which the emergence of real consumer robotic products has happened alongside astounding progress in academic research in the broad field of AI. At the present time, floor cleaners are still the only category of mass-market robot which have achieved significant commercial success. This can be put down simply to the greater difficulty of the other more complex tasks and chores that a consumer might want an autonomous product to achieve. These tasks place much larger demands on a robotic system to understand and interact with its complicated 3D surroundings and the objects they contain. This programme will focus on creating the research breakthroughs needed to enable this next generation capability.

There are scene perception and modelling competences which underly all of these use cases, and these will be our research focus as we develop the algorithms behind next-generation object-based SLAM systems by combining all of our knowledge in state-based estimation and machine learning. We will also work more specifically on the methods for training learning systems; methods for advanced vision-guided manipulation; and the frameworks needed for practical, contextual human-robot interaction. The core scientific work will be forward-looking and academic, but always with a strong guidance from our partners at Dyson.

Planned Impact

Domestic robotics has been forecast to be a major global growth sector (£1B currently to £20B by 2025), and Dyson is well-positioned to be a key driver of this growth with its investments into personnel, research and facilities over the past five years. It is often stated that the UK should aim at leadership in AI, but Dyson is one of the very few UK companies aiming seriously at making that happen at scale with real robot products already on the market and sold worldwide.
Robotic vacuum cleaners and lawnmowers have barely scratched the surface of the possibilities that domestic robotics represent, and the research programme presented in this proposal tackles key fundamental challenges that need to be addressed in order to produce robots that can perform useful functions in the real world alongside humans. Furthermore, specific provisions have been made to ensure a continuous pipeline of technologies from low TRL research at the Laboratory all the way through to high TRL commercial deployment by Dyson and as such Prosperity Partnership funding will allow us to provide a much better flow from technology to industry.

Besides direct wealth creation, the existence of a centre of knowledge and excellence in the field of vision-enabled robotic manipulation will act as a nexus to draw (and retain) much needed skills, capabilities and investment into the UK. The Laboratory already has a good track record of attracting some of the top research talent in the world, and Prosperity Partnership funding will greatly aid in maintaining this attractiveness. The close involvement of Dyson engineers with the Laboratory will provide researchers with grounding in real-world challenges besides giving them a flavour of life in industry should they be interested in pursuing non-academic careers - Dyson most certainly will require many more robotics engineers.
SLAM and its evolution into general robotic spatial awareness remain key areas of interest in the academic disciplines of robotics, computer vision and AI, and we intend to make fundamental and high impact published scientific contributions during the project. Our track record in consistently encouraging and helping students and PDRAs to publish at this level speaks for itself. Doubtless, the research outputs of the Laboratory will also find application in a whole host of different sectors, for example medical robotics, construction, disaster relief, assisted living and manufacturing, all of which require robots that can interact in real-time with complex, dynamic environments.

From a societal point of view, the advent of domestic robots capable of performing a whole range of tasks around the home can be expected to improve the quality of life by reducing the amount of time that is devoted to performing household chores. This will certainly have the largest impact on homemakers and, given that a greater majority of homemakers are still women, the widespread adoption of domestic robotics might also be expected to have beneficial knock-on effects on gender equality issues and household wage demographics. A further corollary to the beneficial effects of domestic robotics is the great potential to extend the independence of an ageing population.

Finally, Dyson and Imperial College have a strong relationship with technical media outlets including The BBC, The Times, The Guardian and Wired Magazine, which will allow us to disseminate our results to a large and worldwide audience. We will also ensure that the key technical demonstrators are showcased at multiple events. Imperial Festival attracts over 5,000 people and generates considerable media interest. Through these engagements with the public, we aim to inspire the next generation of engineers and scientists and generally getting people excited about the potential behind robotics. Lastly, the Lab maintains an open website and this project will have its own space which will be continually updated with the latest developments.

Publications

10 25 50
 
Description Dyson-DRL Partnership 
Organisation Dyson
Country United Kingdom 
Sector Private 
PI Contribution The group is well-known for his work in computer vision and there is transfer of expertise in this area.
Collaborator Contribution Dyson has a long successful history in developing innovative products for home. Their expertise and knowledge in this area enable us to further understand the limitation of computer vision approaches for practical purposes and enable us to direct our research towards more practical solutions.
Impact The Dyson-DRL collaboration under the prosperity partnership award is only a few months old and there are not currently tangible outcomes.
Start Year 2019