http://blog.deathpod3000.com Documenting the construction and programming of an autonomous r/c car Sun, 12 Jul 2009 17:24:24 +0000 http://backend.userland.com/rss092 en This post will show how to calculate bearing and distance to a latitude longitude waypoint using C on an Amtel AVR microcontroller.  Deathpod3000 uses the formulas I gave in the last post, but with minor changes to adapt to C on the AVR.  The code should be easy to adapt ... http://blog.deathpod3000.com/?p=207 In testing code for Deathpod3000, I found it very valuable to have a quick and easy way to calculate correct distance and bearing to a latitude longitude waypoint to compare to results I was getting from my Atmega32 C code.  In this article, I will explain a little about the ... http://blog.deathpod3000.com/?p=192 Someone shot video of 2 of the first round trials at the Sparkfun Autonomous Vehicle Competition including the first run of Deathpod3000.  The video is from a cell phone so it is a little grainy, but it's fun to watch.  I especially like his comments while he's shooting.  :-)  The ... http://blog.deathpod3000.com/?p=181 [caption id="attachment_173" align="aligncenter" width="460" caption="Deathpod3000 Top View"][/caption] There were some requests on the Sparkfun site for the specs of the vehicles so I thought I'd list them.  All the info is in the build posts, but the design changed along the way so here is what we ended up with: (Click ... http://blog.deathpod3000.com/?p=168 Race day finally arrived.  To be honest, I had been spending so much time on the car, I was happy that spending every spare minute on the build was finally over.  The ir sensors that we added in the last post were working well for avoiding obstacles and curbs so ... http://blog.deathpod3000.com/?p=140 After the test run on the actual race course, it was clear that GPS was not going to be accurate enough to keep the car out of the pond.  I tried sonar, which is great for general obstacle avoidance because it will detect basically anything in a 45 degree cone, ... http://blog.deathpod3000.com/?p=134 In the last post we took a run around the race course and found out we will need obstacle avoidance.  I forgot one step - putting the final board together.  In this installment I will show how I read rc server pulses from the receiver allowing me to use the ... http://blog.deathpod3000.com/?p=124 After installing a new gps and compass in previous posts, we took the Deathpod to a parking lot for a test run around a simulated track.  The results were not bad.  The car went around the track at 30% throttle and 50% throttle.  At 50%, it tended to spin out ... http://blog.deathpod3000.com/?p=117 The new 5hz gps module I added in the last post is working.  Now, to tackle the biggest problem discovered during Deathpod3000's maiden voyage, we need a way to get very fast heading updates during turns.  While the faster refresh rate of the new gps will help, it is never ... http://blog.deathpod3000.com/?p=108 So the maiden voyage of Deathpod3000 was a disaster.  The problems seemed to be twofold.  One, gps updates were too slow and two and probably most importantly, heading info coming from the gps was way too slow to be useful in turns.  The net result was that it mostly drove ... http://blog.deathpod3000.com/?p=96