Closed vincent-hui closed 5 years ago
To all, I ran into some hardware issues last week and couldn’t test the branch. I should be able to test the code early this week and send some feedback.
On Nov 5, 2018, at 04:39, Zihan Chen notifications@github.com wrote:
@vincent-hui https://github.com/vincent-hui @linhongbin https://github.com/linhongbin Is this PR ready for review? Thanks.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/jhu-dvrk/dvrk-ros/pull/28#issuecomment-435811889, or mute the thread https://github.com/notifications/unsubscribe-auth/ABEX81LKKg_zlb2rFMXlR3jAq2Xrloqwks5usAc8gaJpZM4XXLUy.
Hi all, I push a new version for matlab codes of dvrk GC. Since the time we push the fisrt version, we have improved our method to achieve a more stable and reproducible gravity compensation process. The 2nd version is not compatible with the 1st one. So after you pull from the remote, you might need to rerun all the process by calling:
[GC_controllers,output_data_struct] = gc_demo('MTML&MTMR')
If you have done data collection using 1st version, the data folder should be decapitated.
For separated testing, you might need to run(takes MTML for example): For datacollection: [is_dataCollection_finish, output_data_struct] = dataCollection('MTML', 'dataCollection_config.json');
For estimation: [is_MTM_lse_finish, output_dynamic_matrix, output_lse_config] = MLSE('MTML', 'MLSE_config.json', './../GC_Data_stable_test')
For GC_controllers: [is_test_pass,GC_Controllers] = GC_controller('MTML','GC_controller_config.json','./../GC_Data_stable_test','4POL',9.81)
The whole data collection process might takes around 1 hour for each arm.
Looking forward for news of review.
Just to confirm, this PR is now obsolete and the update code can be found at https://github.com/jhu-dvrk/dvrk-gravity-compensation
@vincent-hui , @linhongbin , can we close this PR?
yes, we can close this PR.
@vincent-hui @linhongbin Is this PR ready for review? Thanks.