Menu
▾
▴
Re: [simspark-devel] Nao model
|
From: Feng X. <hen...@ma...> - 2009-03-06 03:36:22
|
Hi, I totally agree with you that the physical models are much more important than how a robot looks like. However, "the unrealistic results" are not just because of the models. And I think it is largely caused by the current simulator. Such as, unrealistic servo motor models, the control system via servo speed, ODE can not simulate elastic joint, friction force of the soccer environment and the rigid bodies (robot and ball), noise models, etc. 04:56:16 2009-03-06 Jie Ma: Hi All, From a research perspective physical models are much more important than how a robot look like -- even it may be too late to use them in this years' competition. We've examined the current physical models under the biped locomotion problem, but the current models leaded to very unrealistic results compared with the published work based on the real NAOs. So it would be extremely useful if Aldeberan can help us to establish more realistic physical models. Regards, Jie Sander van Dijk wrote: > Hello all, > > We have contact with the people from Aldeberan, they will supply us with > meshes for the graphical model. In the simulator this model is seperate > from the physical model, so changes in the looks have no effect on the > performance. > > Peter and I have also asked them for some details for the physical model > (mass distributions, sensor noise, etc.). It may be too late to push > them into the simulator for this year, since they will change the > performance and teams may need a lot of time to addept their behaviors > to them, but it will help to make the model more realistic later on. > > Cheers, > Sander > > On Thu, Mar 5, 2009 at 12:08 PM, Daniel Polani <d.p...@he... > <mailto:d.p...@he...>> wrote: > > Hi, > > Aldebaran seemed quite concerned with the appearance of the robot, in > fact more so than on the technical specs. Question to the MC: will a > detailed modeling of the appearance affect simulation performance? > > A second issue is the following: we need to be careful that the > requirements of the simulation league are not necessarily dictated by > the exact technical specifications where this is not yet relevant. Of > course, in the course of years we want to converge to model the actual > Nao, but we need to be careful that our scientific goals are dictated > by what we want and are able to do this year in terms of scientific > challenge. > > - Daniel > > Sahar Asadi writes: > > Hi, > > > > In Suzhou, I talked with one of the guys from Aldebaran and he > gave me his > > contact info. He was positive and said they can provide us more > details for > > improving our model. I wanted to link Feng Xue to him but > meanwhile I became > > so busy and I forgot. I will send you the contact information. I > really > > apologize for forgetting this issue. > > > > > > Best, > > Sahar > > > > On Thu, Mar 5, 2009 at 7:29 AM, Peter Stone <ps...@cs... > <mailto:ps...@cs...>> wrote: > > > > > > > > Hi all, > > > > > > I'll email my contact at Aldebaran to ask how they would like to > > > proceed, and will report back. > > > > > > I'll try to set it up so that there can be direct contact between > > > Aldebaran and the MC. > > > > > > Best, > > > Peter > > > > > > > > > > > > > > On 05/03/2009, at 5:11 AM, Sander van Dijk wrote: > > > > > > > > > Hello MC, OC, TC, executive and trustee, > > > > > > > > > > There is an issue that we have to resolve before the > competitions in > > > > > Graz: the use of the Nao model in the 3D simulation league. > As you > > > > > probably know, in Suzhou there was some controversy about this, > > > > > since the people from Aldeberan were surprised to see the model > > > > > without them being notified. In my contact with them at > that time > > > > > they seemed quite willing to let us use it and to supply > details to > > > > > possibly make the model more realistic, but I think it is still > > > > > necessary to get some sort of formal agreement. We don't > want to > > > > > have a clash again in Graz. > > > > > > > > > > At which level should this be done? At least when the > agreement is > > > > > there, I think it's best that the MC discusses with them > about the > > > > > details? > > > > > > > > I agree it's an important issue, and if I remember correctly, > Joschka > > > > tried to contact them not too long ago, but didn't get a reply (I > > > > don't know the details. I'm pretty sure it wasn't about > getting an > > > > agreement, but about technical issues because Aldebaran was > apparently > > > > interested in contributing to improving the model). > > > > > > > > An agreement is a separate issue, I'm not sure how formal it > needs to > > > > be though. An email from Aldebaran would probably be enough > for now - > > > > probably not too difficult to get, I would hope. > > > > > > > > I know Daniel was part of the discussion last year, so I've > cc'ed him > > > > and Joschka too. > > > > > > > > cheers > > > > Oliver > > > > > > > > -- > > > > Oliver Obst form follows function (Louis > Sullivan). > > > > Fon: +61 2 9325 3278 > http://oliver.obst.eu/ > > > > Autonomous Systems Lab CSIRO ICT Centre > http://www.ict.csiro.au/asl/ > > > > > > > > > > > > > > > > Hi,<br><br>In Suzhou, I talked with one of the guys from > Aldebaran and > > he gave me his contact info. He was positive and said they can > provide > > us more details for improving our model. I wanted to link Feng Xue to > > him but meanwhile I became so busy and I forgot. I will send you the > > contact information. I really apologize for forgetting this > issue.<br> > > <br><br>Best,<br><font color="#888888">Sahar</font><br><br><div > class="gmail_quote">On Thu, Mar 5, 2009 at 7:29 AM, Peter Stone > <span dir="ltr"><<a href="mailto:ps...@cs... > <mailto:ps...@cs...>">ps...@cs... > <mailto:ps...@cs...></a>></span> wrote:<br> > > <blockquote class="gmail_quote" style="border-left: 1px solid > rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br> > > Hi all,<br> > > <br> > > I'll email my contact at Aldebaran to ask how they would like > to<br> > > proceed, and will report back.<br> > > <br> > > I'll try to set it up so that there can be direct contact > between<br> > > Aldebaran and the MC.<br> > > <br> > > Best,<br> > > <font color="#888888"> Peter<br> > > </font><div><div></div><div class="h5"><br> > > <br> > > ><br> > > > On 05/03/2009, at 5:11 AM, Sander van Dijk wrote:<br> > > ><br> > > > > Hello MC, OC, TC, executive and trustee,<br> > > > ><br> > > > > There is an issue that we have to resolve before the > competitions in<br> > > > > Graz: the use of the Nao model in the 3D simulation > league. As you<br> > > > > probably know, in Suzhou there was some controversy > about this,<br> > > > > since the people from Aldeberan were surprised to see > the model<br> > > > > without them being notified. In my contact with them at > that time<br> > > > > they seemed quite willing to let us use it and to > supply details to<br> > > > > possibly make the model more realistic, but I think it > is still<br> > > > > necessary to get some sort of formal agreement. We > don't want to<br> > > > > have a clash again in Graz.<br> > > > ><br> > > > > At which level should this be done? At least when the > agreement is<br> > > > > there, I think it's best that the MC discusses with > them about the<br> > > > > details?<br> > > ><br> > > > I agree it's an important issue, and if I remember > correctly, Joschka<br> > > > tried to contact them not too long ago, but didn't get a > reply (I<br> > > > don't know the details. I'm pretty sure it > wasn't about getting an<br> > > > agreement, but about technical issues because Aldebaran was > apparently<br> > > > interested in contributing to improving the model).<br> > > ><br> > > > An agreement is a separate issue, I'm not sure how > formal it needs to<br> > > > be though. An email from Aldebaran would probably be enough > for now -<br> > > > probably not too difficult to get, I would hope.<br> > > ><br> > > > I know Daniel was part of the discussion last year, so > I've cc'ed him<br> > > > and Joschka too.<br> > > ><br> > > > cheers<br> > > > Oliver<br> > > ><br> > > > --<br> > > > Oliver Obst form follows function (Louis > Sullivan).<br> > > > Fon: +61 2 9325 3278 <a > href="http://oliver.obst.eu/" > target="_blank">http://oliver.obst.eu/</a><br> > > > Autonomous Systems Lab CSIRO ICT Centre <a > href="http://www.ict.csiro.au/asl/" > target="_blank">http://www.ict.csiro.au/asl/</a><br> > > ><br> > > ><br> > > <br> > > </div></div></blockquote></div><br> > > ------------------------------------------------------------------------------ Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise -Strategies to boost innovation and cut costs with open source participation -Receive a $600 discount off the registration fee with the source code: SFAD http://p.sf.net/sfu/XcvMzF8H _______________________________________________ Simspark Generic Physical MAS Simulator simspark-devel mailing list sim...@li... https://lists.sourceforge.net/lists/listinfo/simspark-devel Best Regards, Feng Xue 2009-03-06 |
