Pascal Roth
Pascal Roth
@Mayankm96 quick ping, can we get that merged?
@Mayankm96 @Dhoeller19 @jsmith-bdai ping again about that PR and the discussion above
@Mayankm96 thanks for the feedback, will adjust it accordingly
> Hi @pascal-roth - we have a potential early adopter of this sensor at the institute. Would you say it's ready to go or still needs to address some of...
@Mayankm96 and @Dhoeller19, I have an issue with the indented test of the IMU. I want to compare the output with the IMU Sensor of Isaac Sim for a single...
@Mayankm96 I fixed the test. Currently, we are still failing one, and I am not sure why. All the quantities except the lin_acc_w are correct. This calculated lin_vel_w with the...
@jsmith-bdai, everything has been ready for a while. It is literally just the offset computation for the lin_acc_w that has errors, and I am not sure why. Feel free to...
@jsmith-bdai and @jtigue-bdai we investigated the issue further and tried different solvers. By now, we are pretty convinced that it has to do with Pysx. Raised a new issue which...
the main issue for me is currently still the acceleration errors; even the angular acceleration is really off which then also negatively effects the linear accelerations. In my understanding, the...
yes, if you run the default anymal test now, the angular velocities are as follows: ``` # ang_acc_w imu_link tensor([[-2.4121, -5.4381, -0.0945], [-2.3416, -5.4702, -0.1113]], device='cuda:0') # ang_acc_w base tensor([[-1.0775e+01,...