[Archimedes Talks Series] Integrating Domain Knowledge with ML/AI Efficiently for Complex Systems Design and Operation

Title: Integrating Domain Knowledge with ML/AI Efficiently for Complex Systems Design and Operation

Speaker:  Professor John S. Baras (Institute or Systems Research and Department of Electrical and Computer Engineering University of Maryland College Park, USA)

Abstract: The complexity of systems has increased dramatically. As a result, it is not feasible to construct accurate models for many system components. We now have large datasets and advances in ML and AI that can be used in Systems Science and Engineering problems. We propose an Integrated Data-Driven (ML and AI) and Model-Based Systems Engineering (IDDMBSE) framework for the design of autonomous robotic systems. Such designs for specific use cases require careful co-design of both the hardware and the software elements with multiple domain-specialist teams working on different aspects of the system often without the knowledge of concurrent changes being implemented by a different team. Developing such complex systems can significantly benefit from a formalized process that can provide a unified framework for the design of the system. The Model- Based Systems Engineering (MBSE) framework was introduced precisely to tackle this problem through the use of Systems Modeling Language (SysML) as the unifying tool for modeling and managing requirements, structure, and behavior of the system and guiding the overall design process through design optimization and verification using co-simulation. However, the need for autonomy in robotic systems has resulted in the incorporation of data-driven algorithms and techniques in the autonomy stack that cannot be fully captured in the conventional MBSE framework. Our framework addresses this challenge of designing systems by integrating model-based and data-driven techniques. We have developed a novel methodology and three software tools, PERFECT, TRADE-X and VERITAS towards this end. Our IDDMBSE framework is general and can be applied to many autonomous systems. We demonstrate the theory and the tools in two use cases: robust and safe path planning for autonomous vehicles, safe and robust robotic manipulation tasks.

Short Bio: John S. Baras is a Distinguished University Professor, holding the Lockheed Martin Chair in Systems Engineering, in the Institute for Systems Research (ISR) and the ECE Department at the University of Maryland College Park (UMD). He received his Ph.D. degree in Applied Mathematics from Harvard University, in 1973, and he has been with UMD since then. From 1985 to 1991, he was the Founding Director of the ISR. Since 1992, he has been the Director of the Maryland Center for Hybrid Networks (HYNET), which he co-founded. He is a Fellow of IEEE (Life), SIAM, AAAS, NAI, IFAC, AMS, AIAA, Member of the National Academy of Inventors (NAI) and a Foreign Member of the Royal Swedish Academy of Engineering Sciences (IVA). Major honors and awards include the 1980 George Axelby Award from the IEEE Control Systems Society, the 2006 Leonard Abraham Prize from the IEEE Communications Society, the 2017 IEEE Simon Ramo Medal, the 2017 AACC Richard E. Bellman Control Heritage Award, and the 2018 AIAA Aerospace Communications Award. In 2016 he was inducted in the University of Maryland A. J. Clark School of Engineering Innovation Hall of Fame. In June 2018 he was awarded a Doctorate Honoris Causa by his alma mater the National Technical University of Athens, Greece. His research interests include systems, control, optimization, autonomy, machine learning, artificial intelligence, communication networks, applied mathematics, signal processing and understanding, robotics, computing architectures, formal methods, network security and trust, systems biology, healthcare management, model-based systems engineering. He has been awarded twenty patents and honored with many awards as innovator and leader of economic development.