This page is provided for historical and archival purposes only. While the seminar dates are correct, we offer no guarantee of informational accuracy or link validity. Contact information for the speakers, hosts and seminar committee are certainly out of date.
Manipulation robots are like the mainframes of yesterday. They are usually locked in labs, and people occasionally go to them to do an experiment. We've been trying to make robots more like modern computers, and less like physics experiments. In this vision, robots could be organized like today's networks of distributed workstations. In particular, robot teams at varying scales could cooperate to perform manipulation tasks. Manipulation is the acid test for a robot team, since success or failure is so easy to evaluate.
This vision presents a host of scientific and engineering challenges. To illustrate, I'll describe a team of small autonomous mobile robots that cooperate to move large objects (such as couches). The robots run asynchronous manipulation protocols with no explicit communication. I'll discuss some lessons from this small-scale parallelism, including the role of minimal systems with reduced resource requirements (sensing, communication, computation, and state). I'll argue that quantifying tradeoffs in resource requirements can measure the power gained by distribution.
We have also addressed massively-parallel distributed manipulation. I'll discuss our work on sensorless manipulation using SIMD arrays of microfabricated actuators. I'll describe our progress in building the M-Chip (Manipulation Chip), an array of programmable micro-motion pixels. I'll show a prototype M-Chip containing over 11,000 silicon actuators in one square inch. This may be a kind of record for parallelism and density. SIMD manipulation algorithms for parts-sorting and -orienting are developed and analyzed using our theory of programmable vector fields.
I will discuss what lessons carry over between these different scales of size and parallelism. Videos of all our systems will be shown.
Bruce Donald received the SM and Ph.D. degrees from MIT, working under Tomas Lozano-Perez. He is co-founder of the Robotics and Vision Laboratory at Cornell University, where he is currently an associate professor of Computer Science. Donald has written three books and numerous scientific papers on Robotics; this year, he is on leave at Stanford University and Interval Research Corporation.