The objective of this workshop is to discuss the design, development, and deployment of hardware/software systems for the recognition (from humans) and generation (by the robot) of social behaviors that enable robots to be more effective co-workers with their human counterparts in real-world collaborative manufacturing environments. A strong emphasis will be put on systems that support bidirectional communication between humans and robots for closed-loop operations in human-robot collaborative scenarios. This workshop will bring together academic researchers and industry partners to frame problems and solutions in real-world contexts.

Robotic technologies have become well-established in the manufacturing industry for reducing worker ergonomic stress and workload by performing operations quickly, repeatedly, and accurately. With the introduction of compliant robotic agents (e.g., Rethink Robotics’ Baxter and Sawyer, Kuka’s iiwa, Franka’s Emika, etc.), there is a growing opportunity for collaborative partnerships between humans and industrial robots jointly completing various manufacturing tasks, such as assembly, inspection, box-packing, and part-delivery, among others. As robots continue to be integrated into the workplace as versatile aids for industry automation, it is important to develop effective human-robot communication mechanisms that facilitate seamless cooperation and intuitive communication between humans and robots.

Such collaboration increases manufacturing productivity by effectively combining the capabilities of each partner: the intelligence, experience, and responsiveness of human co-workers, and the accuracy, repeatability, and speed of robotic assistants. A key to facilitate direct and physical human-robot collaboration is developing clear, natural, and bidirectional communication methods to exchange shared understanding and goals between both human and robot collaborators.

In human-human collaborations, workers use a variety of social signals—such as speech, pose, gestures, and touch—to bilaterally transmit commands and task-relevant information. Endowing robots with similar capacities requires new contributions in human-robot interaction (HRI), including gestures, visual attention, proxemics, feedback cues, and turn-taking, among others.

Transferring these signals and collaboration techniques from the lab to industrial application poses novel HRI challenges. While these technologies can contribute to improved ergonomics, increased productivity and worker safety, industrial application is still very difficult as machine safety standards have to be satisfied.