Cybernetic Training for Autonomous Robots: Generalizable Autonomy in Robot Manipulation Towards Human Augmentation

Scott Lab E141
201 W. 19th Avenue
Columbus, OH 43210
United States

Garagic

Speaker: 

Dr. Denis Garagić, Chief Technology Officer, Sarcos Technology and Robotics Corporation

Abstract: Human-worn exoskeleton robots such as the Guardian® XO® exoskeleton and dexterous manipulation-capable mobile platforms such as the Guardian® XT®  system have extensive potential benefits to the commercial and industrial sectors. The wide variety of potential applications of these platforms include: (i) commercial and residential construction; (ii) logistics; (iii) transportation; (iv) energy production; (iv) inspection; (v) manufacturing in the automotive, aerospace, and heavy industries; and (vi) repair and maintenance, to list just a few.

The benefit of dexterous mobile robots can be multiplied by allowing such systems to learn from their operators how to execute tasks and then be able to conduct those tasks on their own in a weakly structured environment. This approach is the province of a platform that we refer to as “CYTARTM” – short for Cybernetic Training for Autonomous Robots.

Existing robots perform poorly in dynamic and unstructured environments. Traditional machine learning (ML) requires a time-consuming and expensive trial-and-error approach. Existing AI/ML solutions are environment-specific and hard to replicate as conditions change. As a result, many industries cannot benefit from the efficiencies offered by technological advancements. Sarcos’ Cybernetic Training for Autonomous Robots (CYTARTM) technology program is focused on bringing autonomy to unstructured workplaces.

The CYTARTM platform is being developed to build on emerging Machine Learning/Artificial Intelligence (ML/AI)-powered robotics technologies to enable repair and maintenance, logistics support and, in the long run, the industrial environment of the future where hundreds of networked exoskeletons (XO® robots)-equipped operators and CYTARTM humanoid robots work as a team to achieve increased productivity and prevent worker injuries.

The talk will provide an overview of robotic systems that combine human intelligence, instincts and judgment with the strength, endurance and precision of machines to (i) augment human performance and awareness and (ii) operate autonomously when trained or remotely controlled by a human operator in unstructured environments.

Bio: Dr. Denis Garagić serves as Chief Technology Officer for Sarcos Robotics. In this role, Dr. Garagić is responsible for technical guidance across all innovation, engineering, and technology teams. Dr. Garagić also oversees overall technology development and utilization plans and plays an integral role in helping to develop the company’s strategic direction, development, and future growth.

Prior to becoming CTO, Dr. Garagić served as the Chief Scientist at Sarcos and was responsible for advancing the Sarcos artificial intelligence (AI) program by developing and implementing innovative control strategies, intelligent algorithms, and machine learning (ML) in future generations of Sarcos innovative robotics solutions.

Dr. Garagić has more than 25 years of experience in AI&ML, Optimization and Controls technologies. Prior to joining Sarcos, he served as Chief Scientist at BAE Systems FAST Labs, guiding the creation of cognitive computing solutions that provided machine intelligence and anticipatory intelligence to solve challenges across the Department of Defense and intelligence community.

He has been a Technical Review Authority, Principal Investigator, or Research Lead on numerous programs, including DARPA and Air Force Research Labs (AFRL) research programs. Dr. Garagić is also a regular speaker at international meetings and conferences on AI & ML, influencing how these technologies are transforming the defense, intelligence, and security industries.

Dr. Garagić has written more than 40 journals, articles, and conference papers and holds several patents in applied controls for autonomous robotics and machine learning. He received his B.S. and M.S. degrees in Mechanical Engineering and Technical Cybernetics from The Czech Technical University in Prague and received his Ph.D. in Mechanical Engineering from The Ohio State University.