Rider Tracking in the Amgen Tour of California
Posted: February 27th, 2007 | 3 Comments »The Amgen Tour of California took place this week. During this year’s edition, 7 riders have been tagged with CSC’s OmniLocation devices to send a constant stream of GPS data to the T-Mobile GSM network (over general packet radio service signals, HTTP streaming and SMS messaging). In consequence, The race’s web spectators had access to almost-real-time location information displayed on Yahoo! maps for the riders with a short delay offset (i.e. about 10 seconds).
Screenshots from the live webcast featuring the rider tracking system.
Relation to my thesis: Using sensors to track performances in sport is nothing new. Yet, this is an example of deployment in a semi-uncontrolled settings (the GSM network provider was a sponsor). Moreover, I am wondering about the web spectator’s experience in watching dots moving (without smooth animation but with icons of riders jumping from one location to another) on a map. Finally, the web application does not seem to take into account the delay offset from the acquisition of the location to the display on the screen. I suspect that a sharp synchronization of the geodata had to be performed prior to the visualization.
Fabien;
I designed the Wireless GPS Object Field Tracking application for CSC and I can tell you it absolutely is a new application of technology. Nobody has ever put three ounce, battery-powered GPS/GSM devices on multiple human beings and tracked them in near real time as they move in a group across open terrain. The introduction of SiRFstarIII technology for seeing more than three GPS satellites meant that trees and mountain shadows did not impact location tracking. The use of newly available Wide Area Augmentation System firmware meant that accuracy had finally improved from +/-15m to +/-3m. Had we done this same application last year the results would have been comical at best.
The visualization that you saw was absolutely raw – no synchronization of data was performed prior to presentation. Smooth animation was impossible with 10 different objects updating on their own 10-15 second schedules. Moreover, given that some riders were willing to carry the 90 gram trackers in their jerseys, while others had them tied under their seats – with very different responses from saddles of differing design.
We had a grand total of five weeks of development, including the time to find, procure, test, reject, an reprocure trackers, firmware, and software application programming interfaces, not to mention writing the application. We had one week to test the devices on bikes at the CSC training camp in Gilroy before the tour began. The fact that the trackers survived 650 miles of pounding without any hardening still amazes me. That Morocco has better GSM coverage than California was also a challenge as the trackers entered and exited GSM blind spots from Monday through Thursday.
We got a ton of responses from viewers from California to Afghanistan. Our site tracker indicated over 34,000 unique users during the week, with very long average sessions. We also got response from industry and the military that indicate they have finally seen evidence that Location Awareness technology has matured sufficiently to be of use.
I believe that in less than 6 months we will be able to show sychronized tracker updates at 1-2 seconds, resulting in smooth animation without error-producing predictive mapping. In less than 12 months we will be able to show biometric data on riders and power output from bikes along with location information. We are also working on universally acceptable placement of waterproof, ruggedized, aerodynamic trackers, on either the handlebar stem or the seat post. Lastly, we are working with the makers of communication technology to put tranceivers into the lead, peloton, and chase vehicles to receive signals from the bikes, consolidate them into a multi-threaded stream, and send them at higher power to radio towers or satellite to eliminate blind spots.
Thanks a lot for your very complete comment, Daniel.
It was not my intend to discredit your work around the Wireless GPS Object Field Tracking application. I have an engineering background and I fully grasp the constraints in setting up such an environment in the wild. However, I need to be critique about the state of the art in order to define the problems my research tries to tackle. Your answer definitively describes some of the issues around the design and deployment of location-aware applications I address in my work. More specifically, I try to understand how and when the limitations in terms of location quality and timeliness makes an application less usable or unacceptable to users in collaborative settings. Moreover, I intend to come up with design solutions to design location-aware application that take into consideration the imperfection of the sensed data and uneven network coverage.
Hi Fabien -
I wrote the Adobe Tour Tracker this year and would love to chat more with you about bringing your work into that project for future Tours. If that is interesting to you, send me a note when you get a chance! Allan Padgett, Adobe Systems