A predictive driver model for the virtual excavator

Hydraulic excavators play an important role for various construction tasks. But due to increasing costs of fossil fuels and an increasing environmental awareness, there is high demand for more efficient machinery. During the development of the necessary technologies, simulations based on an excavator model and measured driving cycles allow for an early assessment of the expected machine performance. However, the outcome of such simulations is not only influenced by the cycle and the machine model, but also by the chosen driver model. In this work, a predictive driver model for the virtual excavator is presented. Taking typical real-world driver behavior into account, the developed driver model is based on an event-based task coordinator and a model predictive controller for each axis. Simulation results indicate that the developed driver model captures the actual driver behavior very well.