Abstract: Methods and systems for joint execution of complex tasks by a human and a robotic system are described herein. In one aspect, a collaborative robotic system includes a payload platform having a loading surface configured to carry a payload shared with a human collaborator. The collaborative robotic system navigates a crowded environment, while sharing a payload with the human collaborator. In another aspect, the collaborative robotic system measures forces in a plane parallel to the loading surface of the payload platform to infer navigational cues from the human collaborator. In some instances, the collaborative robotic system overrides the navigational cues of the human collaborator to avoid collisions between an object in the environment and any of the robotic system, the human collaborator, and the shared payload.

Abstract: Methods and systems for collaboration between two robotic vehicle systems to accurately determine a geometric model of the footprint of a loaded robotic vehicle are described herein. A scanning robot is employed to scan a robotic vehicle loaded with a payload. The scanning robot measures the geometric information required to determine a geometric model of the loaded robotic vehicle. The scanning robot traverses a trajectory around the payload robot, while one or more distance sensors repeatedly measure the distance between the scanning robot and the payload robot and one or more image capture devices repeatedly image the payload robot. A geometric model of the footprint of the payload robot is generated based on the collected image and distance information. In some examples, virtual boundaries are defined around the payload robot based on the geometric model to navigate with obstacle avoidance.