Study of Autnomous Pond Excavation in Real Environment

The control of excavator requires specific skills to operate and is also dangerous in some situation. Substituting the manual operation with automation technology would be essential and convenience for the user. Therefore, an autonomous excavator is very interesting and still an active research area. As testing the autonomous excavator in full-scale is difficult to realize, the autonomous excavator system was developed by integrating sensors to a modified excavator model with scale1:14 in order to study the feasibility and limitation of the autonomous excavator in real environment. The model was actuated by the DC motor and hydraulic cylinder controlled by servo valve. The environment around the excavator was sensed by the LiDAR. The LiDAR was applied to generate an elevation map which was later processed to the pond’s dimensions. The encoder was integrated at the joints of the excavator to identify the kinematic of excavator. The system was operated by the Robot Operating System (ROS). The task of autonomous excavator in this work was to excavate the pond with given dimensions autonomously. Using the data from elevation map and the kinematic of the excavator the control algorithm planed the path of the excavator movement to achieve the desired dimension of the pond. The study showed that the system can excavate the pond autonomously with satisfactory results. The knowledge and some technical issues from this study could be transferred to the development of the real autonomous excavator.