What is a relatively large principal moments of inertia?

[alexdupond]

Hi,

I'm working in a bio-inspired robot which is formed as a stick insect. I scaled up the robot by 5 times, because I got some undesired behavior. Now most of the problems are gone, but I still have some problems with high spikes in force sensor values and the robot exploding.

I think that maybe it is the mass or inertia of the body parts but Im not sure that is meant by relatively large principal moments of inertia? (https://www.coppeliarobotics.com/helpFi ... ations.htm)

For example i have a part of the robot consisting of a cylinder of (bounding box) size:
x = 2.75 * 10^-2
y = 2.75 * 10^-2
z = 2.415 * 10^-1

The dynamic parameters are:
mass = 7.152* 10^-2 kg

inertia X = 7.5 * 10^-4
inertia Y = 3.9 * 10^-2
inertia Z = 3.9 * 10^-2

I don't know if it is possible to make any conclusions of this. Alternatively I can upload my model if needed.

Thanks for your time.

[coppelia]

Hello,

uploading your model is always helpful. What physics engine are you using for the simulation? With Bullet and ODE, you most certainly will have to tweak values to obtain something satisfactory, with Vortex you'll be able to keep quite close to real-world values. Newton also, but it is somewhere in-between.

Cheers

[alexdupond]

Hi again

Sorry for the slow reply. I'm using Vortex engine and it seemed to do the trick just increase the moment of inertia with a factor two. I'm working with a very small robot so maybe very small values can be a problem.

But thanks for the reply I will keep that in mind.

Best regards
Alexander