NASA Aeronautics technical talks at NASA Ames Research Center were broadcasted for free online.
Research, Technology, and Challenges for Safe Self-Driving Car Operations
Abstract: Self-driving cars will transform mobility by making transportation safer and more efficient, and by giving people back the time currently wasted on the manual operation of vehicles. In 2009, Google started a self-driving-car program to rapidly advance autonomous-driving technology. Building on previous work in the field, Google has developed a small fleet of vehicles that to date have autonomously driven over a million kilometers in complex real-world conditions. In this talk, Dr. Dolgov described the history of the project, discuss the research and technology that leads to safe self-driving operations, and talk about the challenges that lie ahead and the future of autonomous driving.
Dr. Dmitri Dolgov leads the software team on the Google Self-Driving-Car project. Prior to that, he worked on self-driving cars at Toyota and at Stanford as part of Stanford's DARPA Urban Challenge team. Dmitri received his B.S. and M.S. in Physics and Math from the Moscow Institute of Physics and Technology in 1998 and 2000, respectively, and his Ph.D. in Computer Science from the University of Michigan in 2006. In 2008, Dmitri was named one of "AI's ten to watch--the Future of AI" by the IEEE Intelligent Systems Magazine.
Sorry. Video not available.
Kiva: Challenges of Autonomy
Abstract: Kiva System’s mobile fulfillment system blends techniques from control systems, machine learning, operations research, and other engineering disciplines into the world’s largest mobile robotic order fulfillment platform. Kiva uses hundreds to thousands of mobile robots to carry inventory shelves around e-commerce fulfillment centers. Mobile inventory enables the optimization of storage and retrieval of physical inventory using techniques developed for computer data, making order fulfillment fast and efficient. Their robots are running in over 50 fulfillment centers on multiple continents for Amazon, Staples, and many others, with new sites opening every year. The first Kiva warehouse opened for Staples in 2005. The original robots have been delivering shelves to help fill customer orders every day since, even as the site has expanded over the years. This talk will discuss what autonomy means at Kiva and will explore some applications of learning for adaptation, challenges in handling failure, and lessons learned over almost a decade of continuous operation.
Andrew Stubbs, Senior Systems Engineer, is responsible for systems architecture at Kiva Systems LLC. Prior to joining Kiva over eight years ago, Andy developed autonomous helicopters at Nascent Technology, created a hovercraft test bed for decentralized control at UIUC, and designed semiconductor equipment at Ulvac. He has a MS/Ph.D. in Mechanical Engineering from the University of Illinois and a BS in Mechanical Engineering from Boston University.
Matthew Verminski, Vice President of Hardware Architecture, is responsible for leading the hardware architecture efforts at Kiva. Prior to joining Kiva over nine years ago, Matt guided an assortment of product development efforts from concept to production at Virtual Ink, Sengine, and PowerAdvocate. He was an inventor of the Virtual Ink's flagship product Mimio, an award-winning digital whiteboard recorder. Matt holds 13 United States and six international technology patents, with several more pending. He has a MS in Electrical Engineering and Computer Science from the Massachusetts Institute of Technology and a BS in Computer Engineering from Tufts University.
Abstract: Self-driving cars have the potential to transform mobility: they will make transportation safer, give freedom to millions of people who can't drive, and give people back their time. A dedicated team at Google has spent the last five years moving self-driving vehicles closer to a reality. New algorithms, increased processing power, innovative sensors and massive amounts of data enable our vehicles to see further, understand more and handle a wide variety of challenging driving scenarios. Our vehicles have driven over a half million miles on highways, suburban and urban streets. In this talk, I will discuss Google's overall approach to solving the driving problem, the capabilities of the car, the company's progress so far, and the remaining challenges to be resolved.
Jiajun Zhu is one of the founding engineers of the Google's self-driving car program. He designed and developed most of its perception system in the early years of the program and is currently the technical lead of the perception and the simulation teams. Prior to joining the self-driving car team, he led several large scale computer vision programs on the Google maps team. He earned his M.Sc. in computer science from the University of Virginia in 2007 and B.Sc. from Fudan University in 2005.
Sorry. Video not available.