Between the last post almost a month ago, the wall-following robot has undergone some significant changes to address deficiencies identified in January
- First and foremost – the use of 4ea AA batteries for power. The plan is to replace this pack with 2ea 3.7V 2000mAH Li-ion batteries from Sparkfun, charged via 2ea Sparkfun PowerCell chargers. Moreover, these will be moved to the bottom of the robot rather than sitting on top. This required a custom-designed bracket to hold the batteries and chargers, but that’s no problemo with my new PowerSpec PRO 3-D printer (well, actually it *was* a problemo, as it took about 6 revisions to get the bracket optimized).
- The current motor driver module is located on the underside of the robot, and its heatsink barely clears the floor. In fact it started dragging the floor on some of the later tests last month, as the double-sided foam tape holding it up began to sag. The plan is to move this unit to the top, into the space vacated by the clunky AA pack.
- More ping sensors: The old unit had 3 – 2 side-looking and one forward-looking. After figuring out the positive feedback problem in the original configuration and turning the unit around, it had 2 side-looking and one rear-looking sensor, so that obviously has to change. In addition to moving the rear sensor to the front, I want to add two additional side-looking sensors directly over the main wheels so they don’t suffer the ‘swing’ effect due to the long moment arm from the turn center to the sensor. My hope is by combining the information from the on-axis and off-axis side-looking sensors, I can improve Wall-E’s wall-following performance.
Here are some preliminary movies – enjoy!