Developing a Self-optimizing Robotic Excavator System with Virtual Prototyping Technology

Nowadays, a robotic excavator is widely applied in several fields of mining industry, agriculture, and forestry. Due to the complexity of dig environment such as slope of terrain profile, hardness of soil and unknown disturbances inside the soil, a versatile dig strategy avoiding obstacles has become a challenge to be addressed. This article presents an engineering model that describes the mechanical behavior of the robot excavators when it interacts with soil and generates an appropriate bucket path to adapt with each soil type, terrain profile and avoids buried obstacles as well. Based on a concept of the digging process, a robot excavator was designed in SOLIDWORKS and then exported to ADAMS environment. Whereas a soil model was modeled and analyzed by finite-element analysis (FEA) in ANSYS, then exported to ADAMS environment. The interactive simulation was implemented in ADAMS environment to investigate the dynamic behavior of robot and soil. Through the self-optimizing behavior of the robot in the digging process, an optimal bucket trajectory and an intelligent control strategy were generated in MATLAB/Simulink to control the robot tracking the desired bucket trajectory. The self-optimizing strategy was evaluated by effectiveness of the proposed algorithm in testing scenarios with many soil types and obstacles on the virtual prototype model.