Can MoveIt create plans for future move group states?
I'm currently working to optimize the cycle time of a repeated set of arm movements. We're using MoveIt to plan and execute trajectories in C++. It can take up to half a second to plan and optimize a trajectory using the MoveGroupInterface::plan function and I'd like to parallelize this since the start state of the next move is known ahead of time if the current move is successful.
Is it possible to specify a start joint state and request a plan from that start state to an end state/pose, and if so how is this done?