[Rock-dev] Change in Eigen 3.2 which breaks the base types (and maybe other packages like orientation estimator)

Alexander Duda Alexander.Duda at dfki.de
Sat Nov 23 19:06:44 CET 2013


Am 23.11.2013 um 17:07 schrieb Javier Hidalgo Carrió <javier.hidalgo_carrio at dfki.de>:

> On 11/22/2013 02:53 PM, Matthias Goldhoorn wrote:
>> I just pushed a update for the base/types/Pose.hpp
>> the implementation of euler changed in eigen:
>> 
>> http://eigen.tuxfamily.org/dox/group__Geometry__Module.html#gad118fececd448d7485ffea4858775e5a
>> 
>> The returned vector is in in form of:  [0:pi]x[0:pi]x[-pi:pi].
>> 
>> before there was no definition for that.
>> so far we in rock assumed that the vector is in: (-pi,pi),
>> (-pi/2,pi/2),(-pi/2,pi/2), since the conversion would be more
>> complicated that the manual calculation i added a new to euler helper
>> method to the pose which does the conversion like the old eigen code (up
>> to 3.1) does.
> I did not fully get the point. What is the problem of using 
> toEulerAngles(a0, a1, a2) from Eigen?
> The new euler helper (base::getEuler) only computes the euler angles in 
> ZYX in this order.
> There are multiple conventions for proper Euler angles.
> 
> Javier.

Yes, we should not limit this and leave it up to the user which conversion he wants to perform. But the new toEulerAngles conversion from eigen validates the ZYX definition used by cars,ships and aircrafts in terms of valid angle ranges (there are always two possible solutions how to present a certain pose in ZYX order).

Alex

> 
>> 
>> But if you call the eigen method within your code manually take respect
>> to this change. The new version is pushed to master only so far.
>> 
>> The testsuite passes...
>> 
>> Matthias
>> 
> 
> 
> _______________________________________________
> Rock-dev mailing list
> Rock-dev at dfki.de
> http://www.dfki.de/mailman/cgi-bin/listinfo/rock-dev



More information about the Rock-dev mailing list