Skeleton-free RULA ergonomic assessment using Kinect sensors

Abstract

Ergonomic analysis methods and postural assessments assist in preventing work related musculoskeletal injuries. The primary outcome of this is to provide safer workplace environments and reduce the total number of work related injury occurrences. The dedicated marker- or sensor-based 3D motion capt ure solutions impose a series of portability and cost challenges when applied to work place environments. The Kinect motion recording system, which was introduced in the Xbox game console, provides a marker-less, portable and low-cost motion capture solution. This technology is capable of recording 3D coordinates of a moving body parts with an accuracy comparable to the state of the art systems. This paper investigates the utilisation of Kinect sensors for real time rapid upper limb assessment (RULA) to aid in ergonomic analysis for assembly operations in industrial environments. Unlike earlier similar attempts, the work presented in this paper does not rely on tracking body parts and extracting a kinematically sound skeleton. In this paper, identifying a RULA score is formulated as a semantic segmentation problem. A random decision forest (RDF) classifier is used to give each pixel a different RULA score based on postures captured with a Kinect camera. Results demonstrate a converging accuracy of 93% for the proposed method.