Capability-Driven Adaptive Task Distribution for Flexible Multi-Human-Multi-Robot (MH-MR) Manufacturing Systems 2020-01-1303
Collaborative robots are more and more used in smart manufacturing because of their capability to work beside and collaborate with human workers. With the deployment of these robots, manufacturing tasks are more inclined to be accomplished by multiple humans and multiple robots (MH-MR) through teaming effort. In such MH-MR collaboration scenarios, the task distribution among the multiple humans and multiple robots is very critical to the efficiency and also more challenging due to the heterogeneity of different agents. Existing approaches in task distribution among multiple agents mostly consider humans with assumed or known capabilities. However, in reality human capabilities are always changing due to various factors, which may lead to suboptimal efficiency. Although some researches have studied several human factors in manufacturing and applied them to adjust the robot task and behaviors. However, the real-time modeling and calculation of multiple human capabilities and real-time adaptive task distribution in flexible MH-MR manufacturing according to human capabilities are still challenging due to the complexity of human capabilities and heterogeneous multi-agent interactions. To address these issues, this paper first proposes a practical modeling approach to model and calculate the capabilities of different humans in real-time using some measurable performance indices. Based on these capabilities, this paper furthermore mathematically models the MH-MR manufacturing process and proposes a capability-driven adaptive task distribution approach with genetic algorithm based solutions to online distribute different tasks to humans and robots. The proposed adaptive approaches are validated through different MH-MR manufacturing tasks and the experimental results show that the approaches can significantly improve the manufacturing efficiency in terms of the time cost and the number of accomplished tasks than existing approaches in the presence of different time-varying human capabilities. Detailed results and statistical comparisons are presented to illustrate the effectiveness and advantages of the proposed solutions.
Shaobo Zhang, Yunyi Jia
Chang’an University, Clemson University