I'm elated to say that my mbed-powered autonomous vehicle, Data Bus, won 3rd place in the 2012 Sparkfun Autonomous Vehicle Competition (AVC)! Thanks, mbed!!
Last Saturday (6/16) was the annual Sparkfun Autonomous Vehicle Competition where ground and air robots run a timed race around the Sparkfun Building in Boulder. It's an awesome event to watch (if you're not stuck in the pits fixing your robot...)
My mbed runs a sensor update, estimation, control routine at 100Hz and about 20,000 lines of code. It's barely taxed. Heading estimation is accomplished with a time-lag compensated Kalman Filter that fuses GPS heading and gyro heading rate data. Top speed was 20mph and 37.16 seconds around the building. Official results:
1st Place: Team 0x27, 2.08 seconds (hoop time deduction; raw 22.08 sec)
2nd Place: Team Minuteman, 9.46 seconds (hoop; raw 39.46 sec)
3rd Place: Team Databus, 37.16 seconds (no hoop)
Top speed of my robot was 20mph with an estimated cross-track error of about 1 meter after the 270m course.
If it wasn't hard enough to build a robot to drive itself 270 meters around the narrow parking lot of a large building, giant red barrels were added on the first leg. A hoop was situated in the center of the path; robots passing through it earn a 30 second time deduction.
But a misstep a few meters later could land a robot off the curb and down into the pond! If the robot makes it this far, on the next leg, curbs jut out waiting to grind unlucky bots to a halt and, surviving this obstacle a giant pothole awaits robots as they approach the last turn.
I counted 34 ground vehicles out of 50 entrants that actually made it to the event. Of those I think something like 7 or 8 made it around the building. As for the air vehicles (doing this in three dimensions isn't just 3 times harder) I have no clue how many made it or didn't...
Sparkfun Recap Teaser
{{http://3.bp.blogspot.com/-pjbPd2pjx-s/T99Uf2CjumI/AAAAAAAAFpM/qNdj3grdXWM/s400/2012_sf_avc_094.jpg}}
I'm elated to say that my mbed-powered autonomous vehicle, Data Bus, won 3rd place in the 2012 Sparkfun Autonomous Vehicle Competition (AVC)! Thanks, mbed!!
Last Saturday (6/16) was the annual Sparkfun Autonomous Vehicle Competition where ground and air robots run a timed race around the Sparkfun Building in Boulder. It's an awesome event to watch (if you're not stuck in the pits fixing your robot...)
My mbed runs a sensor update, estimation, control routine at 100Hz and about 20,000 lines of code. It's barely taxed. Heading estimation is accomplished with a time-lag compensated Kalman Filter that fuses GPS heading and gyro heading rate data. Top speed was 20mph and 37.16 seconds around the building. Official results:
1st Place: Team 0x27, 2.08 seconds (hoop time deduction; raw 22.08 sec)
2nd Place: Team Minuteman, 9.46 seconds (hoop; raw 39.46 sec)
3rd Place: Team Databus, 37.16 seconds (no hoop)
Top speed of my robot was 20mph with an estimated cross-track error of about 1 meter after the 270m course.
{{http://3.bp.blogspot.com/-2Ci6kaofQnA/T99UsLprrqI/AAAAAAAAFpk/5SN43jwkask/s400/2012_sf_avc_108.jpg}}
The ground winner, Team 0x27, 1/5 scale buggy
[[http://susanbrady.smugmug.com/Events/2012-SparkFunAVC/23626357_VmgGB5#!i=1911719671&k=dS2vs63|Pics my friend Susan took]]
[[http://www.bot-thoughts.com/2012/06/2012-autonomous-vehicle-competition-avc.html|Here's a write-up of my experiences Saturday]]
[[http://www.bot-thoughts.com/2011/05/data-bus-nickle-tour.html|More detail about my robot]]
[[http://www.youtube.com/playlist?list=PL890A8244F3EE6DC5|I am collecting AVC videos on this playlist]]
If it wasn't hard enough to build a robot to drive itself 270 meters around the narrow parking lot of a large building, giant red barrels were added on the first leg. A hoop was situated in the center of the path; robots passing through it earn a 30 second time deduction.
{{http://1.bp.blogspot.com/-PQDZ1pGxBHo/T99UmkVWXAI/AAAAAAAAFpc/s1rpnvrEdHs/s400/2012_sf_avc_098.jpg}}
But a misstep a few meters later could land a robot off the curb and down into the pond! If the robot makes it this far, on the next leg, curbs jut out waiting to grind unlucky bots to a halt and, surviving this obstacle a giant pothole awaits robots as they approach the last turn.
{{http://3.bp.blogspot.com/-Eb0SWe2BQcg/T99VUCHNlgI/AAAAAAAAFrc/Emc_XDxrRqg/s400/2012_sf_avc_081.jpg}}
{{http://4.bp.blogspot.com/-50ZMp-LB7Hc/T99U7Uc805I/AAAAAAAAFqc/YhHMhSSftoM/s400/2012_sf_avc_030.jpg}}
{{http://4.bp.blogspot.com/-SzmKiP19YLI/T99VBFGv1PI/AAAAAAAAFqs/qVnQjOGQYAc/s400/2012_sf_avc_041.jpg}}
I counted 34 ground vehicles out of 50 entrants that actually made it to the event. Of those I think something like 7 or 8 made it around the building. As for the air vehicles (doing this in three dimensions isn't just 3 times harder) I have no clue how many made it or didn't...
{{http://3.bp.blogspot.com/-1-dEsiW3Jhs/T99VQ-aob-I/AAAAAAAAFtI/K77sPpJrjOw/s400/2012_sf_avc_075.jpg}}
Sparkfun Recap Teaser
{{http://www.youtube.com/watch?v=yEhUhVX8OqE}}
{{http://www.youtube.com/watch?v=UL2_dbX97KE}}
{{http://www.youtube.com/watch?v=SbsL4IThgnQ}}
Your Data Bus rover is just great. I have printed a copy of your Data Bus Nickel Tour post to help me with my own rover plans and ideas. Also, I keep reading your write-up article in Robot Magazine... very well written and informative. I have learned a lot lately from your experiences, and thanks for making the code and project details all available to others.
As you may know, I am trying to plan/design my own rover with (at present) a Wild Thumper platform. I will likely use a Propeller if I get a chassis kit from Parallax (it comes with it). But I also have an mbed and I can't believe what you have done with this on your Data Bus. Man, I have so many questions, but with time I hope to be able to wrap it all together.
Would the 8 cores of the Propeller offer anything of value over the mbed as a controller for a Wild Thumper rover? I want to use a Parallax GPS, PING sensor with rotating mount, an XBee, and some sort of stand-alone digital camera. Also, I need to incorporate some type of failsafe, like with your RoverMUX. I know you have commented on my post at the Parallax site, but if you have any additional ideas I am open to hear them! Thank you.
George Sutton
georover@outlook.com
Your Data Bus rover is just great. I have printed a copy of your Data Bus Nickel Tour post to help me with my own rover plans and ideas. Also, I keep reading your write-up article in Robot Magazine... very well written and informative. I have learned a lot lately from your experiences, and thanks for making the code and project details all available to others.
As you may know, I am trying to plan/design my own rover with (at present) a Wild Thumper platform. I will likely use a Propeller if I get a chassis kit from Parallax (it comes with it). But I also have an mbed and I can't believe what you have done with this on your Data Bus. Man, I have so many questions, but with time I hope to be able to wrap it all together.
Would the 8 cores of the Propeller offer anything of value over the mbed as a controller for a Wild Thumper rover? I want to use a Parallax GPS, PING sensor with rotating mount, an XBee, and some sort of stand-alone digital camera. Also, I need to incorporate some type of failsafe, like with your RoverMUX. I know you have commented on my post at the Parallax site, but if you have any additional ideas I am open to hear them! Thank you.
George Sutton
georover@outlook.com
Important Information for this Arm website
This site uses cookies to store information on your computer.
By continuing to use our site, you consent to our cookies.
If you are not happy with the use of these cookies, please review our
Cookie Policy
to learn how they can be disabled.
By disabling cookies, some features of the site will not work.
Access Warning
You do not have the correct permissions to perform this operation.
George Sutton.
3
replies
I'm elated to say that my mbed-powered autonomous vehicle, Data Bus, won 3rd place in the 2012 Sparkfun Autonomous Vehicle Competition (AVC)! Thanks, mbed!!
Last Saturday (6/16) was the annual Sparkfun Autonomous Vehicle Competition where ground and air robots run a timed race around the Sparkfun Building in Boulder. It's an awesome event to watch (if you're not stuck in the pits fixing your robot...)
My mbed runs a sensor update, estimation, control routine at 100Hz and about 20,000 lines of code. It's barely taxed. Heading estimation is accomplished with a time-lag compensated Kalman Filter that fuses GPS heading and gyro heading rate data. Top speed was 20mph and 37.16 seconds around the building. Official results:
1st Place: Team 0x27, 2.08 seconds (hoop time deduction; raw 22.08 sec) 2nd Place: Team Minuteman, 9.46 seconds (hoop; raw 39.46 sec) 3rd Place: Team Databus, 37.16 seconds (no hoop)
Top speed of my robot was 20mph with an estimated cross-track error of about 1 meter after the 270m course.
The ground winner, Team 0x27, 1/5 scale buggy
Pics my friend Susan took
Here's a write-up of my experiences Saturday
More detail about my robot
I am collecting AVC videos on this playlist
If it wasn't hard enough to build a robot to drive itself 270 meters around the narrow parking lot of a large building, giant red barrels were added on the first leg. A hoop was situated in the center of the path; robots passing through it earn a 30 second time deduction.
But a misstep a few meters later could land a robot off the curb and down into the pond! If the robot makes it this far, on the next leg, curbs jut out waiting to grind unlucky bots to a halt and, surviving this obstacle a giant pothole awaits robots as they approach the last turn.
I counted 34 ground vehicles out of 50 entrants that actually made it to the event. Of those I think something like 7 or 8 made it around the building. As for the air vehicles (doing this in three dimensions isn't just 3 times harder) I have no clue how many made it or didn't...
Sparkfun Recap Teaser