The overall goal of this project is to create a sustainable workplace where autonomous robots and humans work together safely and efficiently.
Robots can assist humans with heavy or repetitive tasks that might harm humans if carried out over long periods. On the other hand, some tasks that are easy for humans pose a real challenge for robots; for example, picking and mounting cable harnesses is beyond the capabilities of robots.
The potential for human-robot collaboration is vast in industrial applications and many other areas like health, agriculture, and household robotics. However, the recent advances in perception technologies, machine learning, and online situation-aware planning and control open up possibilities for a much high level of interaction between humans and robots.
The interaction between humans and robot systems is minimal today, severely restricting the use of robots in many scenarios where a trained operator might have close interaction with an automated robotic system. However, this requires efficient interaction strategies between the human and robot, a robotic perception system that can estimate the current state of the environment but also make qualified predictions about human intentions.
Finally, a planning and control system during run-time will adapt the system’s current and predicted state to realize control actions that are safe for a trained human operator but also provide sufficient performance.
The project is coordinated by Chalmers University of Technology. Knut Åkesson is the project manager.