NSF CPS PI Meeting 2012 [National Harbor, Maryland]

In early October 2012, I had an exclusive opportunity to NSF's Principal Investigator Meeting in Cyber-Physical Systems, which was held in National Harbor, Maryland (10 miles from DC). It's basically a 3-day research-in-progress conference for professors who got funding from National Science Foundation. I was there purely by chance.

Venue

National Harbor is a very fancy waterfront on the Potomac River just a bit south of DC. 

Gaylord Hotel, front.

This area has exorbitant prices. Even a mid-class hotel network like Hampton charges $250/night before taxes and fees.

Hampton Inn, where I stayed.

The conference was held at the Gaylord Hotel and Convention Center, which is a huge dome with a small park with two-story houses inside it. This complex is impressive both in size and the way it's constructed.

Inside Gaylord.

Inside Gayord, towards Woodrow Wilson Bridge.

Besides several hotels, the area had a few shops and nice restaurants. I personally tried the ones with Thai, American, and Italian foods. I approve.

One building at the shoreline.

(In addition to all sorts of money-spending places, there is a medium hill to the west of National Harbor. That's where I did a hill sprints workout, but it's not related to the main topic.)

A beach sculpture.

CPS PI Meeting

 There was an interesting mix of people attending the conference:

• NSF and other government people
• Control engineers
• Mechanical engineers
• Embedded electronics engineers
• Industry people: automotive (GM, Ford, Toyota, Chrysler), energy, medicine.

This mix is especially interesting from the perspective of a software engineer like myself.

The main conference room.

 The list of general topics addressed at the meeting:

• Classic control and adaptation, extensions.
• Robustness, uncertainty, and noise: graceful degradation 
• Timing in sensing and computation; schedulability 
• CPS security (emerging) 
• Model integration and synthesis  
• Verification and validation: compositional, fuzzy approaches 
• Human-in-the-loop: both removing and adding; accommodating humans; trust 
• Educating CPS engineers: how to teach all needed fields, how to train in formalisms.

Left to right: myself, Akshay, and Prashant who's our co-PI from Toyota.

There was a huge poster session. The one from our project, presented by Akshay, was about compositional verification of heterogeneous models. To find out more, you can stop by the CPS Architectures group and check out our research project progress.

View on poster. Akshay's is on the left.

To summarize, the meeting was very informative for me; both in terms of problems and their solutions.