A Seminar that Might Interest You–Friday 2/22/13–University Crossings-153

The technical details of this may be over the heads of many of us (myself included), but the topic is clearly relevant to the general focus of this course.

---

Dr. Rahul Mangharam “Closing the loop with Cyber-Physical System Modeling”

Cyber-Physical Systems are the next generation of embedded systems with the tight integration of computing, communication and control of "messy" plants. I will describe our recent efforts in modeling for scheduling and control of closed-loop Cyber-Physical Systems across the domains of medical devices, energy-efficient buildings and programmable automotive systems. The design of bug-free and safe medical device software is challenging, especially in complex implantable devices that control and actuate organs whose response is not fully understood. Safety recalls of pacemakers and implantable cardioverter defibrillators between 1990 and 2000 affected over 600,000 devices. Of these, 200,000 or 41%, were due to firmware issues (i.e. software) that continue to increase in frequency. There is currently no formal methodology or open experimental platform to test and verify the correct operation of medical device software within the closed-loop context of the patient. I will describe our efforts to develop the foundations of modeling, synthesis and development of verified medical device software and systems from verified closed-loop models of the pacemaker and the heart. With the goal to develop a tool-chain for certifiable software for medical devices, I will walk through (a) formal modeling of the heart and pacemaker in timed automata, (b) verification of the closed-loop system, (c) automatic model translation from UPPAAL to Stateflow for simulation-based testing, and (d) automatic code generation for platform-level testing of the heart and real pacemakers. More details here. As time permits, I will describe our investigations in energy-efficient building automation in which we coordinate scheduling of controllers for peak power minimization across multiple plants. We will also briefly discuss in-vehicle and networked vehicle-to- vehicle programmable automotive architectures for the future.

Speaker Bio:

Rahul Mangharam is the Stephen J Angello Chair and Assistant Professor in the Dept. of Electrical & Systems Engineering and Dept. of Computer & Information Science at the University of Pennsylvania. He directs the Real-Time and Embedded Systems Lab at Penn. His interests are in real-time scheduling algorithms for networked embedded systems with applications in automotive systems, medical devices and industrial control networks. He received his Ph.D. in Electrical & Computer Engineering from Carnegie Mellon University where he also received his MS and BS in 2007, 2002 and 2000 respectively. In 2002, he was a member of technical staff in the Ultra-Wide Band Wireless Group at Intel Labs. He was an international scholar in the Wireless Systems Group at IMEC, Belgium in 2003. He has worked on ASIC chip design at Marconi Communications (1999) and Gigabit Ethernet at Apple Computer Inc. (2000). Rahul received the 2012 Intel Early Faculty Career Award and was selected by the National Academy of Engineering for the 2012 US Frontiers of Engineering.

The Advance of AI & Neural Networks

Here’s an article from Wired today

How Google Retooled Android With Help From Your Brain

By Robert McMillan

02.18.13

6:30 AM

A picture of the human voice, courtesy the AndroSpectro app. Photo: Ariel Zambelich/Wired

When Google built the latest version of its Android mobile operating system, the web giant made some big changes to the way the OS interprets your voice commands. It installed a voice recognition system based on what’s called a neural network — a computerized learning system that behaves much like the human brain.

For many users, says Vincent Vanhoucke, a Google research scientist who helped steer the effort, the results were dramatic. “It kind of came as a surprise that we could do so much better by just changing the model,” he says.

AI–Long Range Possibilities

For those interested in the longer range issues of Artificial Intelligence you’ll find there are greatly different views held by knowledgeable observers.  At one pole is Ray Kurzweil, recently hired by Google as it’s new “Director of Engineering.”   He believes in the “singularity” and in infinite life for humans.  He has a long history of producing extraordinary devices.

Here’s a short excerpt from an article from  about those who have doubts about Kurzweil’s vision.

“After writing about Ray Kurzweil’s ambitious plan to create a super-intelligent personal assistant in his new job at Google (see “Ray Kurzweil Plans to Create a Mind at Google—and Have it Serve You”), I sent a note to Boris Katz, a researcher in MIT’s Computer Scientist and Artificial Intelligence Lab who’s spent decades trying to give machines the ability to parse the information conveyed through language, to ask him what he makes of the endeavor.

Here’s what Katz has to say about Kurzweil’s new project:

“I certainly agree with Ray that understanding intelligence is a very important project, but I don’t believe that at this point we know enough about how the brain works to be able to build the kind of understanding he says he is interested in into a product.”

Artificial Intelligence (AI)

 

We’re used to computers performing complex tasks.  Most of what we’re used to depends on the designers of the computer program having anticipated every possibility and defining precise ways for the device to perform.  The field of Artificial Intelligence (AI), addresses the extension of these concepts to systems in which we cannot predict every situation that will be encountered.  A robot or autonomous vehicle would be a good example of such a system.

Wikipedia has a good introduction and overview of the field that is relatively up-to-date in this rapidly changing area.

Artificial intelligence (AI) is the intelligence of machines and robots and the branch of computer science that aims to create it. AI textbooks define the field as "the study and design of intelligent agents"^[1] where an intelligent agent is a system that perceives its environment and takes actions that maximize its chances of success.^[2] John McCarthy, who coined the term in 1956,^[3] defines it as "the science and engineering of making intelligent machines."^[4]

A much more extensive discussion can be found in the article “What is Artificial Intelligence” by  the inventor of the term “Artificial Intelligence” John McCarthy - 2007

In the short time that we can spend in this class we’re not going to become truly knowledgeable about AI, but we can give some basic structure and address several of the major uses.  Hopefully the applicability to the Intelligent building concept will be plausible.

An Introduction to AI – KQED at Stamford 2008 – 10:10min

The headings that Wikipedia uses for the topic are:

• Deduction, reasoning, problem solving
• Knowledge representation
• Planning
• Learning
• Natural language processing
• Motion and manipulation
• Perception
• Social intelligence
• Creativity
• General intelligence
  

Major Sub Areas

• Expert Systems – Interview with one of the originators – Edward Feigenbaum – 16:30min
• Neural Networks – Intro by a student giving an overview – 2009 – 7:49min

Current State of the Art

• Scientists See Promise in Deep-Learning Programs – NYT – 11/23/2012 – (you may need to access through the library)
• Where Artificial Intelligence Went Wrong - Critique of AI – Noam Chomsky 11/1/2012
• A Look Back at Predictive Assistants and Lurching Giants – Tom Simonite 12/25/2012 – MIT Technology review –What happened in 2012 to advance usability using AI concepts.

Predictions for the Future

• Sentient machines: the next step in human evolution – Dick Pelletier 1/8/2013 – brief look at current projects working towards that end.
• Better Than Human: Why Robots Will — And Must — Take Our Jobs – Kevin Kelly 12/24/2012 – Wired.com – A very optimistic view of what will happen as AI/Robots develop.