Improved Gait

A project log for Dogbot

A quadrupedal walking robot

Charles GalambosCharles Galambos 07/11/2018 at 19:586 Comments

So after a bit of tinkering with the gait parameters I managed to come up with something that works much better.   It is still a fixed gait animation without any feedback, but it is stable and robust against disturbances.  Next is to add some feedback from an IMU and the torque from the motors.


davedarko wrote 07/12/2018 at 23:59 point


  Are you sure? yes | no

Dennis wrote 07/12/2018 at 23:50 point

Very nice!!!

  Are you sure? yes | no

Josh Cole wrote 07/11/2018 at 21:17 point

Wow this is awesome progress :D nice job!

  Are you sure? yes | no

ꝺeshipu wrote 07/11/2018 at 21:13 point

I see you added pauses to avoid oscillations. But you still move the legs very fast to avoid falling down, so this is not a statically stable gait. I think you would get statically stable if you moved the body away from the leg that is about to be raised before every step — I do that for my robots and it helps a lot.

  Are you sure? yes | no

Charles Galambos wrote 6 days ago point

It is actually doing as you suggest, the pause is it actually it shifting its weight though it isn't obvious from this shot. Have you any experience with dynamic gaits?

  Are you sure? yes | no

ꝺeshipu wrote 5 days ago point

Ah, I see, it's not obvious.

Not really, my robots are very small and simple, and don't (yet) have any kind of sensor feedback. I did manage to make them trot, which counts as a dynamic gait, but that's achieved by simply moving fast enough that it doesn't have the time to fall. No balancing or anything advanced like that.

I know some theory around the ZMP, but I never actually used that in practice. With your motors acting also as force sensors, it seems you should be able to calculate ZMP for your robot.

  Are you sure? yes | no