Today I did a four foot controlled run, it went about two feet, relatively straight, and then the last two feet it started to curve drastically. When it stopped it was probably about eight inches away from the straight line path. My final evaluation response is, just not worth proceeding with this any further.
I think I will have to do a scratch build robot, using the latest technologies and equipment. I will probably use the Sun Computer method, build the robot using off the shelf parts. Some preliminary thoughts, it will probably have to be a robot that has a base of 12-14" dia, with two drive wheels tucked inside. The wheels themselves, maybe 4-6" dia, whichever is more readily available. Some sort of device, front and back, too keep the robot level. the wheel system will to have some kind of encoder system. I am going to give this some more thought and then probably start another thread with what I have in mind.
Bad robot, going off course so much! Ray, you're a good man and I'd like to help you out. I'll take that renegade recalcitrant robot off your hands for $25, free ship of course.
Scratch building is fun, and once you get into encoders and dead reckoning, the world is your oyster! Here's my first big bot, circa 2010. BS2 and relays worked well!
Looks like you're well into things! Mine are just sitting in a closet at the moment because I am in the process of moving. Hopefully I will get to it before winter is up.
The original unit (modified) was customized by a school or university. It's all original inside with the speech and all (no arm). I am using it to model the custom unit I am building out of the "shell" on the right.
I am starting to feel like a yoyo with this, one day I like it, the next day I hate it. Yesterday I took the RB5X out for a drive, using the remote control, when I started the robot from a dead stop, let it run for about four feet, it did a fairly straight run.
Now I am not sure why I am having such a big difference in the two run scenarios. The other run, I had placed the robot next to the wall so I could see and measure how far off the path it got. Maybe there was something on the wood floor at that spot that made the wheel slip and put the robot off the path. Maybe I have to work in some kind of limited slip differential in the wheel assemblies.:-) It looks like I have to address the wheel traction for the robot, with the dc motors there is quite a bit of torque that is involved, maybe during the run one of the motors stops for a brief moment, and then when it starts back up, the restarting torque kicks to the robot of course. I hope that is not a sign that the dc motors are going bad.
The current wheel system uses very slippery wheels with relays that turn them full on or off. There's no speed correction or adjustment. This is one of the primary things I am doing with the blank RB5X. I was originally hoping to replace the stock wheel system with the MMWK from Parallax, however I think they're too big to fit. I may just look for another solution or build something custom.
Comments
I think I will have to do a scratch build robot, using the latest technologies and equipment. I will probably use the Sun Computer method, build the robot using off the shelf parts. Some preliminary thoughts, it will probably have to be a robot that has a base of 12-14" dia, with two drive wheels tucked inside. The wheels themselves, maybe 4-6" dia, whichever is more readily available. Some sort of device, front and back, too keep the robot level. the wheel system will to have some kind of encoder system. I am going to give this some more thought and then probably start another thread with what I have in mind.
Ray
Scratch building is fun, and once you get into encoders and dead reckoning, the world is your oyster! Here's my first big bot, circa 2010. BS2 and relays worked well!
If you are moving to Utah, does that mean you will no longer be working for Parallax?
Ray
The original unit (modified) was customized by a school or university. It's all original inside with the speech and all (no arm). I am using it to model the custom unit I am building out of the "shell" on the right.
And yes, sadly. :depressed:
Not sure there is anything similar in Idaho.
Now I am not sure why I am having such a big difference in the two run scenarios. The other run, I had placed the robot next to the wall so I could see and measure how far off the path it got. Maybe there was something on the wood floor at that spot that made the wheel slip and put the robot off the path. Maybe I have to work in some kind of limited slip differential in the wheel assemblies.:-) It looks like I have to address the wheel traction for the robot, with the dc motors there is quite a bit of torque that is involved, maybe during the run one of the motors stops for a brief moment, and then when it starts back up, the restarting torque kicks to the robot of course. I hope that is not a sign that the dc motors are going bad.
Ray
The current wheel system uses very slippery wheels with relays that turn them full on or off. There's no speed correction or adjustment. This is one of the primary things I am doing with the blank RB5X. I was originally hoping to replace the stock wheel system with the MMWK from Parallax, however I think they're too big to fit. I may just look for another solution or build something custom.