You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The idea is that if for some reason we detect that the fine calibration requires a change, we might be able to use the previous dataset by explicitly applying the same joint offset change in the logged data, so the estimation from wholebodydynamics will be as if the new fine calibration was being used.
It should be easy to implement since we already have a vector that predetermines the joint order we can just mirror that vector and store the offset delta found in the configurations file there
The text was updated successfully, but these errors were encountered:
Note: this is will surely work for the datasets collected on the pole and for the ones requiring a controller it would depend if the robot is still able to do the required motions with the changed joint offsets.
We may use it in the main configuration file only, for no repetition among all experiments we assume the same joint offsets are required for all linked for all experiments belonging to a configuration dataset.
Now we should decide if it goes on the params file or as a parameter to the read experiment function.
The idea is that if for some reason we detect that the fine calibration requires a change, we might be able to use the previous dataset by explicitly applying the same joint offset change in the logged data, so the estimation from wholebodydynamics will be as if the new fine calibration was being used.
It should be easy to implement since we already have a vector that predetermines the joint order we can just mirror that vector and store the offset delta found in the configurations file there
The text was updated successfully, but these errors were encountered: