A new preprint on arXiv addresses a core challenge in physical human-robot interaction: robots must adapt their control behavior quickly and continuously, without pausing what they are doing. The authors argue that current ROS 2 control architectures fall short because they depend on stop-to-switch mechanisms, which interrupt the task whenever a change in control is needed.
The paper proposes a way to achieve sub-millisecond reflexes, allowing the robot to respond to human input or environmental changes almost instantly. This would be a significant improvement over the stop-and-switch approach, which introduces latency and breaks the flow of interaction.
Because the abstract is truncated, the specific method and experimental results are not described in the available text. The claim is limited to the motivation and the proposed outcome: faster, uninterrupted adaptation for safer and more natural physical collaboration.