Phil Koopman, ECE-Silicon Valley Campus - Carnegie Mellon University

Phil Koopman, ECE-Silicon Valley Campus - Carnegie Mellon University

Philip Koopman

Phil Koopman headshotThe Silicon Valley campus welcomes Prof. Philip Koopman of CMU's Department of Electrical & Computer Engineering.

  • Thursday, August 28
  • 3:30 p.m.
  • Silicon Valley campus, Rm 109/110

Remote Attendance

Via Adobe Connect at cmusv.adobeconnect.com/CMUSVSeminar — log in as guest

Software Robustness Testing and Run-Time Monitoring of Autonomous Vehicles

Abstract: Ensuring autonomous system safety is crucial, but traditional analysis and testing techniques are likely to be insufficient, especially for adaptive systems. To help find problems missed by traditional testing, we have built an automated robustness testing tool set to stress-test robotic software, revealing significant safety and functionality problems that were previously unknown to the designers. Because it may be impractical to fully validate autonomous systems at design time, we are investigating the use of run-time monitoring for ensuring at least some aspects of system safety, and have successfully applied that technique to ensure the safety of a large autonomous off-road vehicle. However, even with such tools, significant challenges remain in ensuring autonomous systems are safe, including tolerating environmental uncertainty, validating machine learning, and providing a realistic role for human oversight.

About the Speaker: Philip Koopman  is a professor at Carnegie Mellon University, where he has worked in the broad areas of wearable computers, software robustness, embedded networking, dependable embedded computer systems, and autonomous vehicle safety. Previously, he was a submarine officer in the US Navy, an embedded CPU architect for Harris Semiconductor, and an embedded system researcher at United Technologies. His current research interests focus on safety critical software and Embedded/Cyber-Physical System education. In addition to the usual academic publications, he holds 26 US Patents in embedded computer systems and related technology, and has conducted more than 140 design reviews of industry software projects. He is a senior member of IEEE, senior member of the ACM, and a member of IFIP WG 10.4 on Dependable Computing and Fault Tolerance.  He has affiliations with the Electrical & Computer Engineering Department (ECE), Institute for Software Research (ISR), and National Robotics Engineering Center (NREC).