Closed kevinzakka closed 3 years ago
This corrects an error I made in returning > 0 reward after the task was complete but the robot was still executing the last movements in a plan (e.g. postplace pose, home pose). The reward is now correctly capped at 1.0 for all tasks.
This corrects an error I made in returning > 0 reward after the task was complete but the robot was still executing the last movements in a plan (e.g. postplace pose, home pose). The reward is now correctly capped at 1.0 for all tasks.