Instead of using gcb supported compass demos I thought of something simpler for a self balancing bot.
Someone must have done this so is it a waste of time?
The idea is a tall chassis with motors at the base and hbridge etc.
It would have a pendulum wired to supply and 2 contacts at the top so if the chassis falls forward the pendulum makes contact and moves the bot in the falling direction to counteract it.
Same if it falls the other way. Only one axis to think about. No gyro chips needed. Anyone tried it?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Always wanted to make a self balancing bot with object avoidance.
Got to check out the mpu-6050_6_axis_solutions again as I have one and it worked with gcb ... and someone went to the bother to write the lib. cheers.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
What would stop the pendulum from oscillating back and forth once set in motion? How would inclines and declines be handled? What happens to the pendulum under acceleration/ braking?
I suspect a waste of time. The way to prove or disprove is is build a prototype and test the hypothesis.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Silly idea but a mechanical solution was.
Maybe a spirit level with opto sensors?
Some mercury in a tube ?:)
A horizontal gyroscope?
No, think it needs the 6050_6_axis solution and selecting the correct motors..
geared dc, stepper ? dunno. See how someone else has done it I suppose.
I've seen some youtube balancing robots that were interesting.
The gcb example I tried serprints numbers to the terminal and that was years ago.
I still got the device and maybe try again.
Most video examples usiing 6050 use an arduino.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Instead of using gcb supported compass demos I thought of something simpler for a self balancing bot.
Someone must have done this so is it a waste of time?
The idea is a tall chassis with motors at the base and hbridge etc.
It would have a pendulum wired to supply and 2 contacts at the top so if the chassis falls forward the pendulum makes contact and moves the bot in the falling direction to counteract it.
Same if it falls the other way. Only one axis to think about. No gyro chips needed. Anyone tried it?
Sounds interesting.
Not done this.
Always wanted to make a self balancing bot with object avoidance.
Got to check out the mpu-6050_6_axis_solutions again as I have one and it worked with gcb ... and someone went to the bother to write the lib. cheers.
What would stop the pendulum from oscillating back and forth once set in motion? How would inclines and declines be handled? What happens to the pendulum under acceleration/ braking?
I suspect a waste of time. The way to prove or disprove is is build a prototype and test the hypothesis.
Silly idea but a mechanical solution was.
Maybe a spirit level with opto sensors?
Some mercury in a tube ?:)
A horizontal gyroscope?
No, think it needs the 6050_6_axis solution and selecting the correct motors..
geared dc, stepper ? dunno. See how someone else has done it I suppose.
I've seen some youtube balancing robots that were interesting.
The gcb example I tried serprints numbers to the terminal and that was years ago.
I still got the device and maybe try again.
Most video examples usiing 6050 use an arduino.
I mentioned the idea to a mate who said I would need a TID motor controller so looked it up and it looks difficult. https://www.e3s-conferences.org/articles/e3sconf/pdf/2020/44/e3sconf_icmed2020_01038.pdf