Re: [Sserver-three-d] Some questions about 3D league

[Klaus D. <kla...@fh...>]

Hi Sander, all,

thanks for the detailed information. I have a follow up question:

Sander van Dijk wrote:
> Hello Abtin,
> 
[snip...]
  > Now, it may not always be this way. The joint motors have limited
> maximum velocity and limited maximum torque, which could mean that they 
> take a little while to achieve the target velocity. This can happen for 
> instance when another object (or the ground) is blocking a body part. In 
> these cases you will get changes in joint angles smaller than expected 
> and you will have to find a way to overcome this problem.
> 
May this also be the reason for a joint to move in the opposite 
direction than commanded? I have strange motor behaviors in some (rare) 
cases as in the following example data that I can not explain right now:

We send negative speed and the joint behaves as expected for the first 5 
cycles. But suddenly in step 6 the joint angle is getting bigger even 
though we still send negative angular velocity. In order to compensate 
between desired and current joint angle the speed amount is even 
increased. This is until in step 10 the joint starts to behave 'normal' 
again. Can this be the result of some ground force?

LHipPitch	
step	DesiredAngle	angleAsIs speed
1	27,732038	29,27	-0,768981
2	26,914383	28,39	-0,7378082
3	26,163174	27,51	-0,6734133
4	25,489485	26,74	-0,6252575
5	24,904392	25,97	-0,53280354
6	24,41897	27,87	-1,7255154
7	24,044292	29,35	-2,652854
8	23,791433	31,35	-3,7792835
9	23,671467	33,11	-4,719267
10	23,695469	31,95	-4,127266
11	23,87389	28,09	-2,108055

Thanks for your help.

Greetings
Klaus

[snip...]