How to dynamically adjust the target trajectory

[The132]

I am using moveit to do a deep reinforcement learning task. The output of DRL is the pose of the end-effector, and its frequency is 10Hz. This means that before the last action is completed, the next action will come. If you simply change the target pose of moveit and execute it, the speed of the robot will be discontinuous, and then there will be frequent jitter. How can I dynamically interrupt the last action that may be executing, let the robot execute the currently planned trajectory and ensure that the speed is continuous？

[welcome[bot]]

Thanks for reporting an issue. Because we're a volunteer community, providing a pull request with suggested changes is always welcomed.

[emielke12]

This may be of help.