Operation Pattern Analysis and Operator State Assessment for Construction Machinery

Aging workforces and labor shortages in the construction industry have created a growing demand for technologies that improve productivity and operational safety. Achieving this goal requires an understanding of operator states, including skill level and workload, which in turn calls for methods that can objectively characterize operational patterns from machinery data. In this research, we apply the time-series modeling techniques developed through our biosignal analysis to hydraulic excavator operation data, proposing a latent-space analysis method based on a variational autoencoder (VAE) that captures operational patterns across multiple temporal scales. Multi-scale analysis with short- and long-time windows enables simultaneous characterization of instantaneous operational features and work-process-level consistency. Metrics derived from the latent space let us capture how operational stability differs between skilled and unskilled operators, and how operation changes as work is repeated. Moreover, because these latent representations can be related to physical quantities such as the bucket-tip trajectory, we consider that representations obtained statistically from data reflect structure grounded in the actual motion of the machine. Once operator states such as skill level and workload can be treated as objective measures, we expect this to become a foundational technology for supporting operator training, monitoring operator state during work, and ultimately designing assist control that adapts to the state of the operator. Technical approaches: - Multi-scale latent-space analysis of operation signals using generative models - Assessment of operational stability and skill level from latent-space metrics - Interpretation of latent representations through correspondence with physical quantities of machine motion