(Too) High number of intermediate points in compute_cartesian_path()
Hey!
I started to work with compute_cartesian_path (currently on Indigo) to move a UR5. As a testcase, I create a list of 5 poses starting at the current pose along the x axis with a distance of 10cm. (The robot should move half a meter). My call looks like this:
self.move_group.compute_cartesian_path(waypoints=pose_list, eef_step=0.02, jump_threshold=10, avoid_collisions=False)
The trajectory looks good, but my plan consists of 46 points (len(plan.joint_trajectory.points)). The docu says "Configurations are computed for every eef_step meters;" so I expect around 25 intermediate points.
This number depends strongly on the jump_threshold. If I set it to a value of less than 1.09, I get a list of 26 intermediate points. For every value above 1.09, I get the all 46 points. The maximal euclidean distance between the joint values in the shorter list however is 0.08 (and 0.04 for the longer list)
Does someone know how the jump_threshold is used and how it is computed?
Nikolas
