[IGSTK-Users] Question about the "Navigator" example

[Patrick Cheng]

Hi Shuiling,

What you are doing is correct.

IGSTK simplified the coordinate system transform calculations. Given
that the underlying scene graph is connected correctly, you can get the
transform between any two coordinate system by calling:

fromCoordinateSys->RequestComputeTransformTo( toCoordinateSys );

In your case, you don't need to know the t_Raw and t_Reference
explicitly to compose transforms along the chain. You can get the direct
transform using the 'RequestComputeTransformTo' method.

Take a look at this wiki page on how to connect igstkSceneGraph.


http://public.kitware.com/IGSTKWIKI/index.php/Needle_Biopsy#How_to_connect_the_scene_graph


Patrick

On 7/23/2010 6:55 AM, ling Shui wrote:
> Hi all,
> 
> Sorry to interrupt you!
> 
> Now I want to use the example "Navigator" in IGSTK to connect to the 
> Robot. Before surgery, entry point and target point is defined in the 
> image coordinate system, then they are transformed to the robot, and 
> these two points must be defined in the robot coordinate system. I'm 
> confused by the transformation relationship. Who can help me to solve 
> this problem? Thank you advance for any help!
> 
> The transformation relationship is showed in the figure of the accessory 
> of the email. My original idea is: Use tool1 (must be calibrated) as the 
> registration tool. After registration, the transform, T_registration can 
> be obtained from the landmark registration and it defines transformation 
> from the reference tool to the image coordinate system (Is it right?). 
> After registration, Tool1 will not be used again, the tracked tool will 
> be changed as Tool2, which fixed on the robot arm. T_cal denotes the 
> calibration transform of Tool2, T_tool2robot denotes the transform from 
> Tool2’s tool tip to robot coordinate system. From page-139 of the IGSTK 
> book, T_raw and T_reference transform can be got from the tracking 
> device (by the way, I use the N DI vicra system as the tracking device).
> 
>  From the figure, it seems that the following formulation is correct:
> 
> T_image2robot *T_tool2robot ^-1 *T_cal *T_raw *T_reference ^-1 
> *T_registration = I
> 
> So, the transform T_image2robot can be calculated, and the two points, 
> entry point and target point also can be calculated from the image 
> coordinate system to the robot coordinate system.
> 
> If my idea is right, how to obtain the transform T_raw and T_refercence 
> in IGSTK? If it is wrong, who can give me any advice?
> 
> Thanks & Regards,
> 
> Shuiling
> 
> 
> ------------------------------------------------------------------------
> Your E-mail and More On-the-Go. Get Windows Live Hotmail Free. Sign up 
> now. <https://signup.live.com/signup.aspx?id=60969>
> 
> 
> 
> _______________________________________________
> Powered by www.kitware.com
> 
> Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html
> 
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/cgi-bin/mailman/listinfo/igstk-users