In the quest to make Humanoid Robots that interact more “naturally” with their human “partners”, researchers involved in the EU-funded JAST Project are well on their way to achieving this goal. Combining the fields of Robotics, Psychology and Cognitive Science, the project makes use of studies in Human-Human collaboration to better understand the qualities that would make Humanoid Robots more “human”. In a world where a better quality of life is high on the list for an ageing population, home care through intelligent, well mannered, Humanoid Robots could become a competitive solution for those in need of assistance.
So how are the people at JAST helping our Robot companions learn to anticipate our needs? Basing their approach around the concept of mirror neurons and our ability to observe and copy the actions of others, these researchers are incorporating what they have learnt about “resonance” and “joint action”, in the neural architecture of their robot. In their own words…
“Our robot has a neural architecture that mimics the resonance processing that our human studies showed take place during joint actions,” says Erlhagen. “The link between the human psychology, experimentation and the robotics is very close. Joint action has not been addressed by other robotics projects, which may have developed ways to predict motor movements, but not decisions or intentions. JAST deals with prediction at a much higher cognitive level.”
Predictive Robotics could be the key to providing adequate care in the home for the housebound and chronically ill. As well as the ability to assist in meal preparation, mobility and the timely and correct administration of medication, intelligent Home-Help Humanoid Robots can call for assistance if their sensors detect abnormal behaviour. Going beyond therapeutic companions and single-task appliances, a multi-tasking personal assistant robot, that learns and collaborates, could open avenues for use in other situations too.
Teamwork, collaboration and the ability to adapt to a variety of complex situations are skills that come easily to Humans. If we are to develop Robots that become competent members of these teams, the study of Neuroscience, Psychology and Behaviour Prediction in humans, seems to be a good place to start. Beyond improving our scientific understanding in the areas of Human Cognition, Neurobiology and Psycholinguistics, the JAST Project could have immense implications for real-world, Human-Robot collaboration. Darpa, for one, has been eagerly awaiting the development of militarily useful Humanoid Robots. These Robots would essentially act as artificial squad members and carry out the most dangerous tasks, currently performed by Human soldiers.
Search and Rescue, Fire, Police and Medical rapid response teams would also benefit from Robotic team members that provide “beyond human” strength and sensing capabilities. The ability to navigate our human-centric environments with uneven terrain, stairs and doorhandles makes socially intelligent Humanoid Robots the ideal design choice for Human-Robot collaborative teamwork. Add to this scenario, the latest developments in technologically enabled Telepathic Communication and sub-vocalisation techniques and the vision of a fully integrated, human-like, responsive artificial entity comes closer to fruition.
EUCogII is a European network for researchers in Artificial Cognitive Systems, Interaction and Robotics that is funded by the Information and Communication Technologies division of the European Commission. The network provides a platform for addressing the challenges of the discipline, the current fragmentation of research and “the lack of a clear agenda”. Their overall goal is to create “a close-knit community of well-trained researchers” that work in unison towards the vision of socially interactive Humanoid Robots. Through conferences, workshops and multi-disciplinary collaboration, members and non-members of EUCogII, form valuable, interconnected partnerships that share knowledge across the traditional boundaries of science. Essentially, it enables scientists to “lift their heads” from their highly specialised areas of research to view the vast panorama of Global Robotics Technology and explore new horizons in virgin territory.
It’s not all theory though and suitable hardware platforms that encourage rapid prototyping and hands-on, “learning by building” have been developed as robotics “toolkits” for developers. One such platform is the European RobotCub open source project with its iCub Humanoid Robot that successfully demonstrates the powerful nature and flexibility of collaborative, open-source projects, on a global scale. In the near future, there is no doubt that progressive institutions will continue to provide the resources for the further advancement of Robotics Technology. The next generation of fledgling researchers must have access to the best tools and collective knowledge if they are to realise the dream of Humanoid Robots that anticipate your next move.
Tags: Artificial Cognitive Systems, EU-funded JAST Project, EUCogII, Global Robotics Technology, humanoid robotics, iCub Humanoid Robot, mirror neurons, neural architecture
