In Robomagellan, we joke that our successes and failures are defined by Freddie Mercury moments or Picard face palm moments. A Picard moment would be like saying “Why aren’t we getting camera data?” and then 5 minutes later realizing the camera is unplugged. A rare Freddie moment occurs after a moment of monumental success, such as Monty (Our robot) making about 8 run-throughs of a simple GPS course in almost exactly the same way. Today was definitely a Freddie Mercury filled day, with many breakthroughs and many successful moments. The Robomagellan team met for 6 hours, and although we are very close to the competition date and are working under pressure we still have time to enjoy what we are doing! A good mix of Picard moments and Freddie moments really makes finishing a project an feel like a true accomplishment.
Today we spent the majority of our time either outside on the blacktop running Monty or working through the steering algorithms.
At first, Monty seemed to sporadically turn with no direction in mind. After searching through the code we found that our calculations of the steering angle was off by a few thousand degrees… Using a crazy little thing called
bug testing, we fixed our calculations (after a few face palm moments) and sent our death on four wheels off on a GPS adventure. Monty’s steering behavior improved, but he was still not breaking free and finding any GPS points. Monty was turning too much and made it impossible to controllably steer towards a point. To fix this, we divided the steering angle by five, thus reducing the sharpness of the turns. The subsequent run was full of slow turns and giant circles but Monty found a GPS point!
Monty started slowing down at this point and although he seemed to scream “Don’t stop me now” we replaced his batteries, upped the speed and then retried the course. It was at this point that we realized full batteries make a robots world go fast; Monty completed the fastest run yet at 1:02 seconds, but made a few harrowing turns at high speeds.
This definitely was an incredibly productive day (and we felt that another challenge bit the dust) but we still have a lot ahead of us. Monty still is flying blind with no sonar or camera code tested yet and his bohemian affinity for poles and other dangerous objects (such as buildings, walls, people, bicycle races, etc) poses a threat to long term survivability. Tomorrow we plan on working on the camera, sonars, obstacle avoidance, and general usability; Currently it takes us a few minutes to set up Monty and get him started, but we need to be able to be able to start in a few seconds to maximize run time come competition. In the event that we cannot overcome our difficulties, some dynamite with laser beams will be enough to defeat our competition.