During IREX 2011, I had the opportunity to check out the Omni-Crawler robot developed at Osaka University. Conceptually, it's pretty amazing. It can 'turn on a dime', or more correctly, it doesn't need to turn at all. The unique Omni-Ball drive enables it to move in any direction in its plane of operation, and can make those moves almost instantaneously.
The Omni-Crawler approach will definitely be a significant benefit in some applications that can be improved by it's capabilities, and some applications that were previously impossible. At the same time, the overall complexity of the design and implementation, at least in the research lab prototype stage, raises some questions about how effectively it could be commercialized. The robot has tremendous potential, if it can be production engineered to become a reliable, cost effective subsystem.