Development of an end-to-end hardware and software pipeline for affordable and feasible ergonomics assessment in the automotive industry

TL;DR

This paper presents a low-cost, modular hardware-software pipeline for automating ergonomic assessments in industrial settings, utilizing open-source tools and custom sensors to improve worker health and safety.

Contribution

It introduces a novel end-to-end system integrating custom IMU sensors and open-source software for ergonomic assessment, ensuring affordability and interoperability.

Findings

Achieved high correlation (0.95) with gold standard joint angle measurements.

Global RULA score difference was less than 5% compared to standard methods.

Demonstrated effectiveness in an automotive factory environment.

Abstract

An end-to-end hardware-software pipeline is introduced to automatize ergonomics assessment in industrial workplaces. The proposed modular solution can interoperate with commercial systems throughout the ergonomics assessment phases involved in the process. The pipeline includes custom-designed Inertial Measurement Unit (IMU) sensors, two real-time worker movement acquisition tools, inverse kinematics processing and Rapid Upper Limb Assessment (RULA) report generation. It is based on free tools such as Unity3D and OpenSim to avoid the problems derived from using proprietary technologies, such as security decisions being made under "black box" conditions. Experiments were conducted in an automotive factory in a workplace with WMSDs risk among workers. The proposed solution obtained comparable results to a gold standard solution, reaching measured joint angles a 0.95 cross-correlation and a Root Mean Square Error (RMSE) lower than 10 for elbows and 12 for shoulders between both systems. In addition, the global RULA score difference is lower than 5% between both systems. This work provides a low-cost solution for WMSDs risk assessment in the workplace to reduce musculoskeletal disorders and associated sick leave in industry, impacting the health of workers in the long term. Our study can ease further research and popularize the use of wearable systems for ergonomics analysis allowing these workplace prevention systems to reach different industrial environments.