In industrial scenarios requiring human-robot collaboration, the understanding between the human operator and his/her robot co-worker is paramount. On one side the robot has to detect human intentions, and on the other side the human needs to be aware of what is happening during the collaborative task. In this paper, we address the first issue by predicting human behaviour through a new recursive Bayesian classifier exploiting head and hand tracking data. Human awareness is tackled by endowing the human with a vibrotactile ring that sends acknowledgements to the user during critical phases of the collaborative task. The proposed solution has been assessed in a human-robot collaboration scenario and we found that adding haptic feedback is particularly helpful to improve the performance when the human-robot cooperation task is performed by non-skilled subjects. We believe that predicting operator's intention and equipping him/her with wearable interfaces able to give information about the prediction reliability, are essential features to improve performance in human-robot collaboration in industrial environments.

No abstract

No abstract

In human-robot interaction frameworks maximizing the team efficiency is crucial. However, it is also essential to mitigate the physical and cognitive workload experienced by the shop-floor worker during the collaborative task. In this chapter we first investigate the impact of the robot interaction role (whether being leader or follower during cooperation) on both the human physiological stress and production rate. Based on that, a game-theoretic approach is proposed to model the trade-off between the maximization of the human performance and the minimization of the human cognitive stress. Then, we describe a closed-loop robot control strategy that, based on the proposed game-theoretic model, enables the robot to simultaneously minimize the human cognitive stress and maximize his/her performance during cooperation, by adjusting its role. Eventually, a real-time task allocation strategy is proposed to both ensure the minimization of the human physical fatigue and the effectiveness of the production process. This method relies on a new sophisticated musculoskeletal model of the human upper-body. All these methodologies have been experimentally tested in realistic human-robot collaborative scenarios involving several volunteers and the ABB IRB 14000 dual-arm “YuMi" collaborative robot.

The separation distance between humans and robots in manufacturing shop-floors has been progressively reduced, thanks to the modern safety functionalities available in robot controllers. However, the activation of these safety criteria usually stops the production or reduces the productivity of machines and robots. With the aim of improving this situation, this paper presents a real-time trajectory optimisation method for collaborative robots. The robot trajectory is parameterised at instruction level, i.e. through the parameters characterizing the robot motion instruction. A genetic algorithm randomly modifies the parameters of the nominal trajectory of the robot, thus obtaining new sets of candidate trajectories. Each trajectory is simulated on a digital twin of the collaborative workspace, which allows to reproduce and simulate the robot motion, and to represent the volume of the work-cell occupied by the human operator. A lexicographic optimization is used to evaluate online the optimal robot trajectory that simultaneously minimizes the risk of collision with the human operator and the trajectory traversal time. The method is validated in an industrial scenario involving the ABB YuMi dual-arm robot for a small parts assembly task.