Human-machine posture prediction and working efficiency evaluation of virtual human using radial basis function neural network

A method using radial basis function neural network (RBF-NN) to calculate the virtual human working posture and ergonomics efficiency in human-machine system is proposed. Two RBF-NNs with appropriate structures are respectively constructed and trained by taking advantage of practical data to quantificationally predict both work-related posture and its working efficiency at any moment during the riveting process in aircraft assembly. The results show that the proposed method provides a reference method in ergonomics simulation and assessment leading to a better design of work.