Quantcast

Inverse Reachability Base Pose

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Inverse Reachability Base Pose

akapusta
I am using Inverse Reachability on the built-in PR2 model to find base positions where the robot can reach some grasp.
The implementation is very standard, straight from the tutorial:

        densityfn,samplerfn,bounds = self.irmodel.computeBaseDistribution(Tgrasp)

The description says: "samplerfn: gaussian kernel sampler function taking number of sample and weight, returns robot base poses and joint states"

I then get what is described as base poses from the sampler function.

        poses,jointstate = samplerfn(N-len(goals))

I then find that each pose is a 1x7 array, instead of a 4x4 transform I might expect, or a 1x3 with x,y, theta. What are the 7 values in the array? 1x7 sounds suspiciously like values for the 7dof arm... The tutorial runs robot.SetTransform(pose) on that 1x7 base pose. I previously have only ever run robot.SetTransform on a 4x4 homogeneous transform. Can it take this 1x7 pose? I do not understand what is going on.

Thanks,

Ari
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Inverse Reachability Base Pose

baxelrod
I have found that sometimes OpenRave uses a 7 element array to represent a quaternion and translation like so: [qw, qx, qy, qz, x, y, z]

this is used for example when adding axes to the scene:
originaxes = misc.DrawAxes(env, [1, 0, 0, 0, 0, 0, 0], dist=1, linewidth=2)


-----Original Message-----
From: akapusta [mailto:[hidden email]]
Sent: Thursday, May 29, 2014 4:55 PM
To: [hidden email]
Subject: [OpenRAVE-users] Inverse Reachability Base Pose

I am using Inverse Reachability on the built-in PR2 model to find base positions where the robot can reach some grasp.
The implementation is very standard, straight from the tutorial:

        densityfn,samplerfn,bounds =
self.irmodel.computeBaseDistribution(Tgrasp)

The description says: "samplerfn: gaussian kernel sampler function taking number of sample and weight, returns robot base poses and joint states"

I then get what is described as base poses from the sampler function.

        poses,jointstate = samplerfn(N-len(goals))

I then find that each pose is a 1x7 array, instead of a 4x4 transform I might expect, or a 1x3 with x,y, theta. What are the 7 values in the array?
1x7 sounds suspiciously like values for the 7dof arm... The tutorial runs
robot.SetTransform(pose) on that 1x7 base pose. I previously have only ever run robot.SetTransform on a 4x4 homogeneous transform. Can it take this 1x7 pose? I do not understand what is going on.

Thanks,

Ari



--
View this message in context: http://openrave-users-list.185357.n3.nabble.com/Inverse-Reachability-Base-Pose-tp4027011.html
Sent from the OpenRAVE Users List mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Time is money. Stop wasting it! Get your web API in 5 minutes.
www.restlet.com/download
http://p.sf.net/sfu/restlet
_______________________________________________
Openrave-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openrave-users



------------------------------------------------------------------------------
Time is money. Stop wasting it! Get your web API in 5 minutes.
www.restlet.com/download
http://p.sf.net/sfu/restlet
_______________________________________________
Openrave-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openrave-users
Loading...