Ezra's Round Table / Systems Seminar: Edward Tunstel (UTRC) - Achieving Next-Level Robotic Intelligence with Systems Engineering Implications

Description

Recent advances have been effective for progressing the state of robotics toward its next level of intelligence. At this stage, further advancement is needed with sharper focus on increased robustness, cognitive facilities, and more sophisticated behavior. Research necessarily continues to focus on the study of intelligent robotics toward enabling an intelligence pipeline including abilities to perceive, reason, plan, act, control and learn. Advances to date are impacting individual elements of this intelligence pipeline but the paths to effective integration of these elements that will realize practical intelligent robots and robotic systems are unclear. Thinking more broadly about intelligent robots as elements of larger systems, robotics technology is riding the same tsunami as AI, machine learning, and Internet of Things — all shaping up as elements within systems of systems. Robots will be nodes in IoT within intelligent homes and buildings including hotels and the workplace, in logistics environments, smart factories, and more. In many applications they will be dynamic physical nodes in a variety of cyber-physical systems. Beyond integration of technologies, additionally considering the human element brings the question of transdisciplinary solutions into the conversation. Robots with next-level intelligence will challenge the tools and methodologies being developed to engineer SoS in a number of ways. This talk offers thoughts and considerations for active researchers to ponder and appreciate in the context of their research programs. From a perspective on current stages of technology development, and looking toward the progress horizon from that vista, the talk discusses a number of robotic intelligence topics using examples from research projects and deployed systems. Aspects of intelligent robotics associated with different domains and applications are discussed such as planetary robotics, disaster response, and wearable robotics. Also addressed are future considerations for the evolution of related technology toward increased robotic autonomy and advanced intelligence for robotic systems. Motivating this discourse are multiple considerations for next-level robotic intelligence technology, including the need to realize smart behavior for not only singular robots but for multi-robot and human-collaborative systems, and some of the related systems engineering considerations. The aim is to broaden the thinking of current researchers leading to the collective leverage that will boost robotic intelligence to the next level enabling robots that are multi-functional in the real-world. Bio: Dr. Edward Tunstel is an Associate Director of Robotics at United Technologies Research Center leading a Robotics Group focused on human-collaborative robotics spanning manufacturing and service to after-market autonomous systems capabilities for the aerospace and defense sectors. He was a Senior Roboticist at Johns Hopkins APL researching robotic intelligence/autonomy, modular open systems, and human-collaborative robotics for national security and space applications, and also served as Space Robotics & Autonomous Control Lead. He was previously with NASA JPL as a Senior Robotics Engineer, Mars rover systems engineer for autonomous navigation, and rover engineering team lead for mobility and robotic arm operations on the surface of Mars. He maintains expertise in robotic autonomy and intelligent systems, in which he has authored over 160 publications. He is an IEEE Fellow, IEEE SMC Society Past President, and IEEE Robotics and Automation Society member holding a Ph.D. in electrical engineering from University of New Mexico and prior mechanical engineering degrees from Howard University.