A bridge too far. It's over.

As hinted on several blogs, we are not semi-finalist to the urban Challenge. We did not make that bridge. Thank you for all the good wishers. At some time, we'll try to write and talk more about our system of ominidirectional cameras using compressed sensing and dimensionality reduction in policies to accomplish what we intended to accomplish during the site visit.

hold pattern

FYI, We are not reporting much as we are preparing for the site visit and trying to make some innovative use of light bulbs :-)
[update June 21: they will be on the rooftop]

Site visit

Our site visit will take place on the afternoon of June 27th at the the Southwest Research Institute. As for the technical report, I am going to hold onto it for a little while since we are still investigating some of the final algorithms/hardware to be implemented for the site visit. I thought we would be done by then but day-jobs and other considerations are making this a little optimistic.

We have a site visit.

See y'all you in San Antonio. The 53 teams remaining are listed here.

Unveiling our algorithm

These entries are our part of our current series on unveiling our algorithm as featured in this robots.net story.

• Obstacle avoidance - Isolating important cues
• DARPA Urban Challenge: Unveiling our algorithm, Part Three, Building the Motor/Perceptual models
• DARPA Urban Challenge: Unveiling our algorithm, Part Deux, Sensor and Actuator basics.
• Small summary of what DARPA's Urban Challenge entails
• DARPA Urban Challenge: The basics
• DARPA Urban Challenge: Unveiling our algorithm

Technical Paper

The technical paper will be submitted to DARPA before June 1st at 5:00 ET. We will make it available on this site at that time. It consists of 25% of the grade given to the entry for a spot to the semi-finals.

Site visit announcement tomorrow

We just received this:

An email with the site visit announcement and press release will be emailed on Friday, May 11 at 1100 AM ET.

DARPA Urban Challenge

on a different note, we were featured on robots.net

Site visit course RDNF and our algorithm presentation

The folks at Southwest Research Institute have done a great job at defining the course for the site visit. CP#6 is where the obstacle (yellow car lies). With respect to unveiling our algorithm, we are in the middle of devising a lot of things (a lot of us have finals) for the time being but we have given an outline on the basics needed to a implement a GPS waypoint following vehicle by pointing out to the effort of a vehicle prepared by a large University next door (disclaimer: we have no connection what so ever with this effort) which eventually did not enter the Grand Challenge. Most entries on the subject will be at http://nuit-blanche.blogspot.com/search/label/darpa%20urban%20challenge
.

Unveiling our algorithm, trade-offs and implementation

It is still a work in progress for some parts, but we are going to unveil the algorithm driving our vehicle: Pegasus Bridge 1, as we go along. We are still gathering data and evaluating them against different algorithms. We expect to talk about the implementation as well as the general concepts from GPS waypoint following, road following, situational awareness and learning rules. We hope to be able to do this in a timely manner, but as we are under-manned, between reporting on the algorithm and making it better we will always go for the latter. The reporting will take place here.

Part 4 Video submission completed.

We sent our video to DARPA yesterday. Next stop San Antonio at the Southwest Research Institute.