Closed pattacini closed 5 years ago
I've replaced the use of device drivers with direct connections to encoders ports. This way, we are able to replay experiments by dumping motor state ports.
https://github.com/robotology/assistive-rehab/pull/213/commits/413e681afe941eb258ff958fc20e790e2f88c0ac fixed the problem we spotted on the misalignment above!
This PR introduces the
robotSkeletonPublisher
module that is responsible for publishing a skeleton to OPC and the viewer, which represents the R1 limbs configuration.In the animation below there's a significant misalignment in the viewer that I suppose might depend on my system. Therefore, a careful test needs to be done to verify the correctness of the computation.
I've made now
Ipopt
,VTK
,YARP::rosmsg
andcer
strict dependencies for the project.@vvasco we might want to have two different modalities for the analysis of the movements:
Finally, for replaying data, this module cannot connect to streams recorded from encoders ports since it requires the presence of the robot (using device drivers). Anyway, we can still dump the OPC content at a suitable rate. However, in this respect, we lack the availability of a tool capable of displaying skeletons read from the OPC stream.