Legacy Forum: Preserving Nearly 20 Years of Community History - A Time Capsule of Discussions, Memories, and Shared Experiences.

Bioloid Humanoid falling on its back

Bioloid robot kit from Korean company Robotis; CM5 controller block, AX12 servos..
4 postsPage 1 of 1
4 postsPage 1 of 1

Bioloid Humanoid falling on its back

Post by stamanol » Wed Feb 25, 2009 11:35 am

Post by stamanol
Wed Feb 25, 2009 11:35 am

Hi,

I have built the humanoid confuguration and when I let it walk with the program from the cd (DemoExample(Humanoid)) it starts walking fine but after 10 steps or so it falls on its back as if it cannot balance its weight. Whats wrong? Is my assembly wrong? Is it something with the software adjustment of the servos,or is it generally a faulty program? How can I correct this behaviour?
Hi,

I have built the humanoid confuguration and when I let it walk with the program from the cd (DemoExample(Humanoid)) it starts walking fine but after 10 steps or so it falls on its back as if it cannot balance its weight. Whats wrong? Is my assembly wrong? Is it something with the software adjustment of the servos,or is it generally a faulty program? How can I correct this behaviour?
stamanol
Robot Builder
Robot Builder
Posts: 9
Joined: Thu Jan 15, 2009 9:56 pm

Post by StuartL » Thu Feb 26, 2009 5:06 pm

Post by StuartL
Thu Feb 26, 2009 5:06 pm

Are the screws tight enough to prevent slop/movement in the joints while it's in motion?

Have you also calibrated the standing stance with the motion editor?

The other problem could just be an uneven floor. Static/precoded motion sequences are very sensitive to uneven surfaces as they're entirely reliant on the centre of gravity being exactly where it was when they were programmed.
Are the screws tight enough to prevent slop/movement in the joints while it's in motion?

Have you also calibrated the standing stance with the motion editor?

The other problem could just be an uneven floor. Static/precoded motion sequences are very sensitive to uneven surfaces as they're entirely reliant on the centre of gravity being exactly where it was when they were programmed.
StuartL
Savvy Roboteer
Savvy Roboteer
Posts: 350
Joined: Mon Jun 04, 2007 3:46 pm
Location: Thatcham, Berkshire, UK

Post by stamanol » Thu Feb 26, 2009 5:54 pm

Post by stamanol
Thu Feb 26, 2009 5:54 pm

You are absolutely right,some screws of the servo horn in one knee joint were very loose.After tightening them I could not reproduce the problem. This thing is too sensitive with its center of gravity I guess.

I havent calibrated it. Is it necessary? It seems to be walking fine now.Is this documented in the manual?
You are absolutely right,some screws of the servo horn in one knee joint were very loose.After tightening them I could not reproduce the problem. This thing is too sensitive with its center of gravity I guess.

I havent calibrated it. Is it necessary? It seems to be walking fine now.Is this documented in the manual?
stamanol
Robot Builder
Robot Builder
Posts: 9
Joined: Thu Jan 15, 2009 9:56 pm

Post by StuartL » Fri Feb 27, 2009 10:07 pm

Post by StuartL
Fri Feb 27, 2009 10:07 pm

stamanol wrote:I havent calibrated it. Is it necessary? It seems to be walking fine now.Is this documented in the manual?


It's not absolutely necessary but centering all of the servos is likely to be slightly helpful in walking on a greater variety of surfaces. You may, also, have lucked out and managed to hit a good setup straight out of the box. I managed to fluke this too and didn't need to recalibrate. Others have had to spend hours doing so.

Try the different walking motions and experiment on a variety of surfaces. If he's stable on a level surface then don't fix what isn't broken :D
stamanol wrote:I havent calibrated it. Is it necessary? It seems to be walking fine now.Is this documented in the manual?


It's not absolutely necessary but centering all of the servos is likely to be slightly helpful in walking on a greater variety of surfaces. You may, also, have lucked out and managed to hit a good setup straight out of the box. I managed to fluke this too and didn't need to recalibrate. Others have had to spend hours doing so.

Try the different walking motions and experiment on a variety of surfaces. If he's stable on a level surface then don't fix what isn't broken :D
StuartL
Savvy Roboteer
Savvy Roboteer
Posts: 350
Joined: Mon Jun 04, 2007 3:46 pm
Location: Thatcham, Berkshire, UK


4 postsPage 1 of 1
4 postsPage 1 of 1