Mitsubishi Electric’s behavioural-analysis AI decodes manual tasks without data training

Mitsubishi Electric shared in a recent press statement that it has developed a behavioural-analysis AI that analyzes the efficiency of manual tasks performed on production sites in just a few minutes without requiring operators to prepare AI-training data in advance, by adopting a probabilistic generative model.

The company shares that this new addition to its Maisart AI lineup quickly analyzes videos of people performing repetitive tasks and then visualizes how tasks can be performed more efficiently for greater productivity. This is, it says, is believed to be the world’s first application of a probabilistic generative model that modelled cyclical (repetitive) physical actions performed during factory work.

The technology, first announced by Mitsubishi Electric on February 13, 2019, has been shown in tests to reduce the time conventionally required for work analysis by up to 99 percent.
Mitsubishi is planning on commercializing the system for the fiscal year ending in March 2026, or later.