Framework and System Models of an Intelligent Excavator

This paper presents an intelligent excavator developed in our laboratory, which is composed of two control levels for robotic excavation. The manual pilot hydraulic control system of a mini excavator is replaced by electro-hydraulic servo control system. The intelligent monitoring system and low-level control system based on CAN Bus is constructed. The nonlinear model of the electro-hydraulic servo control system, the kinematic and dynamic models of the working mechanism, and the soil-tool interaction force model are developed. A vision system with a monocular vision sensor and a Binocular vision sensor is also fitted on the excavator for environment sensing and vision-based control, and the state-machine based upper-level task planning strategy is discussed. A ground leveling experiment is implemented to demonstrate the feasibility and benefit of robotic excavation. The system architecture and models proposed in this paper provide a reference for design of intelligent excavator and other filed robotics.