CMLabs has been involved in a number of projects – this page describes a few interesting ones.
ASIMO Cognitive Map
The project involved the design and evaluation of communication middleware in a distributed humanoid robot architecture. Distributed architectures for implementing tasks on humanoid robots is a design challenge, both in theory and practice. Although important functionality resides within the component modules of the system, the performance of the middleware – the software for mediating information between modules – is critical to overall system performance. The Cognitive Map was
implemented in Psyclone, a framework for constructing large AI systems, and allows sharing and transformation of information streams dynamically between modules. It was implemented and tested on the Honda Motor Corporation’s ASIMO humanoid robot. To identify trade-offs and understand performance limitations in robots with distributed system architectures, we performed a variety of tests on these subsystems under different network conditions, operating systems and computational loads. The results indicate that delays due to our middleware is negligible compared to computational costs associated with actual processing within the modules, provided a network with high enough bandwidth. The Cognitive Map appears to be scalable to an increasing number of connected modules with negligible degradation of package delays.
A robotic agent, built by an international team lead by researchers at Reykjavik University in Iceland, is pushing the boundaries of artificial intelligence by automatically learning socio-communicative skills. It’s a big step towards the ultimate goal of creating intelligence that is both self-sufficient and adaptable in a wide variety of environments.
A research team in Iceland has been leading an international group working on a project, called HUMANOBS, that in 2009 set out to build an intelligent system that can learn socio-communicative skills by observing and imitating humans in social situations. The 3.5-year project was recently completed – but according to Principal Investigators Dr. Kristinn R. Thórisson and Eric Nivel at RU’s CADIA AI laboratory, the results have exceeded even the team’s own expectations.
The experiment CoCoMaps uses an expanded version of the existing Cognitive Map Architecture implemented on Honda’s ASIMO robot in an environment with more complex tasks than already attempted. This will allow the robot to interact in more complex ways, in particular, to simultaneously interact with another robot and more than one person at a time. Thus, the project aims for a group of 2 robots and 2 humans. These systems will enable social interactions that can coexist with the robots’ attention to – and completion of – practical tasks in the workplace. A particular focus is on human detection and tracking algorithms and on an improved dialogue system.
The principal components of the dialogue system, as targeted in this demonstration, have been validated piece-wise in laboratory settings, and some subsystems have been demonstrated to work in combination; a final unified, fully integrated whole remains to be demonstrated and is thus targeted here. Deployment in a single robot has been demonstrated, but demonstration on multiple robots has only been demonstrated in simulation; here targeted for integrated demonstration.