How can we be sure of embedded AI systems and what are the next steps for their applications? Learn about embedded AI during this invited talk with Alessandro Pinto, hosted by assistant professor Siddharth Srivastava.
Invited Talk: Embedded AI
Wednesday, October 3, 2018
Brickyard Engineering (BYENG) 210, Tempe campus [map]
Artificial Intelligence finds applications in a wide variety of domains, from voice and image recognition to personalized advertising, healthcare, ratings and recommendations. Many of these applications are deployed in the cloud and users make choices or judge the quality of the information provided by the AI system.
In this talk, Pinto will focus on a different class of applications: where AI is embedded in a cyber-physical system and is required to make decisions autonomously. First, Pinto introduces a reference architecture for autonomy including situation awareness, decision making and run-time verification.
Pinto also discusses the problem of planning missions in a large environment and present a solution based on a hierarchical planning system which has been used both in aerial and ground vehicles. Moving to a different application, he will discuss high-level view of the distributed system in development which automates the loading and unloading processes of cargo aircraft.
About the Speaker
Alessandro Pinto is a technical fellow at the United Technologies Research Center, working in areas of embedded systems, design and verification methods and AI. Pinto has been recognized for his contribution to autonomous flight and in the area of autonomy.
He has published more than 45 peer-reviewed technical articles on network synthesis, hybrid systems, model-based design methodologies and tools and autonomous systems. He holds a doctorate degree from the University of California, Berkeley and a Laura degree from the University of Rome “La Sapienza.”