Abstract: An ironic twist to Autonomous Systems research over the past thirty-five years has been the inclusion of more humans in the loop. In fact, since at least the times of Aristotle, humans have been intrigued by the idea of creating automatons in their image to escape the bondage of labor. Hence, the goal of the first factory robots — some sixty years ago – was to create ‘lights-out factories’ that operated rigidly-timed assembly lines without human involvement. Yet, almost as old as the desire to create subservient automatons, has been the fear of rebellion of our own intelligent machines. Humans, it seems, want to conflate the near-mystic unpredictability of human emotions with the logical toil of their mechanical progeny. So, it may not be such a surprise that the current pinnacle of autonomy research is moving toward machines that seamlessly work alongside untrained human beneficiaries in unstructured and chaotic environments.

This talk explores that evolution in Autonomous Systems research from standalone machines to machines that interact increasingly with humans of greater technical naivete. From robotic assembly to self-driving cars, to emergency response, and robotic surgery, we are gradually pushing the boundaries of the state-of-the art toward more human involvement with increasingly intelligent machines in less-structured situations. Within this space, the emergency response research community made a determined effort to attack highly-unstructured environments in such difficult scenarios that human cooperative control was assumed to be a necessity. The 9/11 attack on the World trade Center in the US was the first use of robots in a real disaster scenario and almost no autonomy was allowed on-site. In that case, the search for victims was so difficult that roughly 60% of victims found were only discovered in post-analysis of video footage, rather than in real-time. Search in deconstructed environments is difficult because a priori model-based information is mostly unusable and dust and debris makes recognition unreliable. In early attempts, teleoperation was the only viable solution with humans lives potentially in the balance. The DARPA Robotics Challenge later brought researchers from around the world to incorporate higher degrees of autonomy and offline and online simulation to the augmentation of both machines and humans. That evolution has continued into the unstructured world of battlefield robotic surgery, in which virtualized reality fuses teleoperation with full autonomy to allow machines to learn from human experts to support full autonomy in times of crisis.

Bio: Dr. Voyles, the Daniel C. Lewis Professor of the Polytechnic, received a B.S. in Electrical Engineering from Purdue University in 1983, an M.S. in Manufacturing Systems Engineering from Dept. of Mechanical Engineering at Stanford University in 1989, and the Ph.D. in Robotics from the School of Computer Science at Carnegie Mellon University in 1997. He is currently Professor of Engineering Technology at Purdue University and an IEEE Fellow. He was a tenured faculty member at the University of Minnesota from 1997 to 2007 and at the University of Denver from 2006 – 2013. He served as lead Program Director for the National Robotics Initiative at NSF and was a co-founder of the NSF Innovation Corps program. He also served as Assistant Director of Robotics and Cyber-Physical Systems at the White House Office of Science and Technology Policy.
Dr. Voyles’ research interests are in the areas of robotics and artificial intelligence. Specifically, he is interested in the development of small, resource-constrained robots and robot teams for urban search and rescue and surveillance. Dr. Voyles has additional expertise in sensors and sensor calibration, particularly haptic and force sensors, real-time control, and Form + Function 4D Printing.
Dr. Voyles’ industrial experience includes Dart Controls, IBM Corp., Intergrated Systems, Inc., and Avanti Optics. He has also served on the boards of various start-ups and non-profit groups.