"There are several strands of research in the field; I survey six: (1) attempts to avoid planning; (2) the design of flexible plan notations; (3) theories of time-constrained planning; (4) planning by projecting and repairing faulty plans; (5) motion planning; and (6) the learning of optimal behaviors from reinforcements."
... even "simple" human actions are extremely difficult to replicate in robots. Now, MIT computer scientists are tackling the problem with a hierarchical, progressive algorithm that has the potential to greatly reduce the computational cost associated with performing complex actions.
Leslie Kaelbling, the Panasonic Professor of Computer Science and Engineering, and Tomás Lozano-Pérez, the School of Engineering Professor of Teaching Excellence and co-director of MIT’s Center for Robotics, outline their approach in a paper titled "Hierarchical Task and Motion Planning in the Now," which they presented at the IEEE Conference on Robotics and Automation earlier this month in Shanghai.
Over the past few years, Willow Garage has taught its PR2 robot to be autonomous. The Silicon Valley startup showed the bot how to plug itself in. They taught it to navigate on its own. They even taught it how to call for help in case of emergency. On December 8th it was time to see how well those lessons were learned. Willow Garage told the PR2 to start rolling and not stop. It traveled over 70 km in 7 days then it kept going for six days more! On December 21st the bot ended its marathon having gone a total of 138.9 km (~ 86 miles). The PR2’s epic journey probably doesn’t look like much from the outside, after all the robot was simply wandering around the Willow Garage office, but it represents a major accomplishment for the company and for open source robotics. In the future, robots like the PR2 will be able to perform a wide variety of jobs, and without human supervision.
Project Cyborg's capabilities are currently being put to the test by several institutions across a range of fields, from architectural researchers at MIT to biological scientists at Harvard. Skylar Tibbits, an architect who heads up the Self-Assembly Lab at MIT, has been working on models for structures that can assemble themselves – from nano-scale up to buildings.
The way in which the locust's distinctive visual system could be transferred into technology for state of the art vehicle collision sensors, surveillance technology and video games has been detailed as part of robotics research being carried out by an academic from the University of Lincoln (UK).