URDF/RobotModel + joint states to tf robot states in a function call?
Is there a way to get all the functionality of robot_state_publisher in a function call? I want to be able to get transforms within a robot model given hypothetical joint positions in a fast loop, so it can't listen to joint state topics or publish onto /tf.
robot_model.init(urdf_from_robot_description_rosparam_or_xml_string)
transforms = robot_model.update(new_joint_states)
... put transforms into a BufferCore and then can do tf lookups
I was hoping I could do this in python maybe with urdf_parser_py, but if not I'll use C++ instead and refactor robot_state_publisher so I can provide joint states directly and return transforms instead of having a listener and publisher. But it would be nice to save that effort- does moveit provide something similar without going through any planning or IK, just jumps to set angles?
https://answers.ros.org/question/3789... is maybe asking half of the same question but specific to moveit, https://answers.ros.org/question/2501... is also similar
https://answers.ros.org/answers/20916... says it isn't possible (as of 2015)