Researchers at the University of Washington have developed a system that can monitor factory and warehouse workers and tell them if they risk musculoskeletal injuries in real time, and they’re using their knowledge to develop an app.
The researchers used machine learning in the system, which divides activities such as lifting a box off a high shelf, carrying it to a table and setting it down into individual actions and then calculates a risk score associated with each action.
Musculoskeletal disorders happen when workers use awkward postures or perform repeated tasks, both of which can strain the body over time.
The team trained the algorithm on a dataset of videos of people doing 17 activities common in warehouses or factories.
To train and test the algorithm, the team created a dataset containing 20 three-minute 3D videos of people doing 17 activities that are common in warehouses or factories. the algorithm computed risk scores based on what was happening to the participants’ joints.
The team is now working on an app that can monitor workers in real time, providing warnings for risky actions and alerts for high-risk actions.
The researchers published their results in IEEE Robotics and Automation Letters.