Alan Winfield is Professor of Electronic Engineering and Director of the Science Communication Unit at the University of the West of England, Bristol. He conducts research in swarm robotics in the Bristol Robotics Laboratory (BRL)
and is especially interested in robots as working models of life, evolution, intelligence and culture. Alan is passionate about communicating science and technology. He holds an EPSRC Senior Media Fellowship with the theme Intelligent Robots in Science and Society.
In his talk to RPEC on the afternoon of 9th March Professor Alan Winfield gave us an overview of the Bristol Robotics Laboratory and some of its main areas of work. The laboratory was set up at the University of Western England (UWE) in 1993 by the speaker. In 2005 it became a joint UWE/Bristol University Project and is currently housed on the UWE Coldharbour Lane Campus. It is the largest robotics laboratory in the country. The Laboratory is multi-disciplinary in the widest sense, and includes not only electronic, mechanical, software and computer engineers but also biological, psychological and behavioural scientists to mention just some.
An intelligent robot is a self contained artificial device that is able to sense its environment and purposefully act within or on that environment. This is essentially a closed loop function. As an example of a simple robot Alan described Solabot
. This simple robot has two motor driven wheels and two separate solar panels. The left motor derives its power from the right mounted solar panel and the right-hand motor derives its power from the left mounted solar panel. Placed in a maze lighted such that maze walls cast shadows Solabot will find its way through the maze because it steers away from the walls by avoiding the shadows. This simple behaviour based robot requires no processing power and no need to model its environment as it interacts directly with the real world.
The talk moved on to describe work being done by the laboratory on:-
A. Biological Robots
are being used to carry out research into tactical sensors, neuroscience research, and energy autonomy. Such research has possible applications in search and rescue, environment monitoring and in crop harvesting, horticulture, gardening and pesticide free pest control robots. By way of examples Alan described two robots Scratchbot
Scratchbot is a robot that "sees" the world through a set of plastic whiskers which work like a rats whiskers down to the extent of having an electronic copy of the area of a rats brain associated with the whisker function. The brain function has been implemented in conjunction with a group working on rat neuroscience based at Sheffield University and provides confirmation of the groups understanding of how the brain works. It also provides a vehicle for exploring ways of manoeuvring in areas where visibility is restricted due to smoke, dust or lighting failure.
B. Humanoid Robots
Ecobot II is a collaboration with biochemists. It has no batteries but gets it energy from digesting dead flies using bacterial fuel cells. The Ecobot II uses eight fuel cells each containing one fly each. This enables it to operate for up to two weeks although the energy generated is small and has to be accumulated to allow it to make intermittent sorties. The fuel cells are preloaded with a fly and a bacterial charge and sealed; over a period of operation the waste products produced eventually poison the cell. Ecobot III avoids this problem by having a complete artificial digestive system. It catches flies using a pheromone trap and digests the fly like a pitcher plant eventually disposing of the waste. It operates in tank moving to a drinking fountain to drink and a litter tray to dispose of waste. Its electronics have been designed to require extremely low power.
- CHRIS the Co-operative Human Robot Interaction Systems project addresses the fundamental issues which will enable safe Human/ Robot Interaction (HRI). A long list of potential application areas is easy to come by including robots for the care of the elderly and sick, domestic servants, tour guides, hotel porters, 'assistant' robots on construction sites, a raft of military applications, leisure/gaming robots and so on. It seems very likely that robotic devices will be a pervasive element of our future society; there are many indications that this will be a huge opportunity for life enhancement and commercial exploitation. Specifically this project addresses the problem of a human and a robot performing co-operative tasks in a co-located space. These issues include communication of a shared goal (verbally and through gesture), perception and understanding of intention (from dextrous and gross movements), cognition necessary for interaction, and active and passive compliance. A diverse set of disciplines have been brought together to realise an inter-disciplinary solution. The starting point for understanding cooperative cognition will be from the basic building blocks of initial interactions. Engineering principles of safe movement and dexterity will be explored, and developed with principles of language, communication and decisional action planning where the robot reasons explicitly with its human partner. Integration of cognition for safe co-operation in the same physical space will provide significant advancement in the area, and a step towards service robots in society. By way of example Alan described three CHRIS sub-Projects BERT 2, Android Heads,
and Heart Robot
Robotic Torso - BERT2 (Bristol Elumotion Robotic Torso 2): BRL are currently developing an advanced robot torso featuring an expressive digital head, torque sensors, artificial 'skin' (that can "feel") and agile limbs. At the same time they are investigating novel adaptive control algorithms for safe human robot interaction. A first draft of the design can be seen here.
C. Swarm Bots
Android Heads These have been used to advance the modelling and generation of realistic, dynamic facial behaviour in humanoid robots. Unlike most research projects in this field, the focus is on dynamic, subtle, facial expressions, rather than static exaggerated facial displays. BRL have investigated ways of capturing human facial motion from video for the animation of robotic faces.
Realistic, life-like robot appearance is crucial for sophisticated face-to-face robot/human interaction. Robot appearance and behaviour need to be well matched to human equivalent responses in order to meet our expectations, formed from our social experience. Violation of these expectations, due to subtle imperfections or imbalance between appearance and behaviour results in discomfort in humans that perceive or observe the robot. Japanese roboticist Masahiro Mori described this in his theory of "The Uncanny Valley", published in the early 70's.
Researchers predict that one day, robotic companions will work with, or assist, humans in space, care, education and many other fields. The effects of the "Uncanny Valley" would be counterproductive to efforts to achieve trustworthiness, reliability and emotional intelligence. All these are basic requirements for robotic companions, assisting astronauts in space or care robots employed as social companions for the elderly or infants.
The Heart Robot project is funded by an EPSRC grant and seeks to explore the human response to robots. The project took robotics out of the lab and into the street to meet people and get them thinking about how we might change when the machines around us seem to have emotions. Will we get better at dealing with such machines, or will we get emotionally fatigued when our gadgets use the language of emotions to help them to communicate with us?
Swarm robotics is a new approach to coordinate the behaviours of large number of relatively simple robots in decentralised manner. As the robots in the swarm have only local perception and very limited local communication abilities, one of the challenges in designing swarm robotic systems with desired collective behaviour is to understand the effect of individual behaviour on the group performance. Alan described the SYMBRION Project as an illustration of the work being carried out in this area
Symbiotic Evolutionary Robot Organisms (SYMBRION)
BRL is a partner in Symbrion, a five-year project being led by the University of Stuttgart, which is investigating the principles of how large swarms of robots can evolve and adapt together into different organisms based on bio-inspired approaches.
The aim is to investigate and develop novel principles of behaviour, adaptation and learning for self-assembling robot "organisms" based on artificial evolution and evolutionary computational approaches. The plan is to combine bio-inspired evolutionary paradigms with robot embodiment and swarm-emergent phenomena thus enabling the "organism" to autonomously manage its own hardware and software organization. It is hoped that such artificial organisms will become self-configuring, self-healing, self-optimizing and self-protecting from hardware and software points of view. This may lead not only to extremely adaptive, evolvable and scalable robotic systems, but might also enable the robot organisms to reprogram themselves without human supervision; to develop their own cognitive structures and, finally, to allow new functionality to emerge: the most suitable for the given situation.
RPEC would like to thank Professor Alan Winfield for finding time in his very busy schedule to come and give us a very interesting and thought provoking talk. Further details of the projects touched on in this review and Information on other BRL projects can be found here at www.brl.ac.uk