Abstract
This paper analyzes the rate of work in robotic harvesting with a head-feeding combine robot. The robot plans a target path for each entire operation, follows the path based on the navigation data from a Global Navigation Satellite System (GNSS) receiver and a GPS compass, and unloads the harvested grain into a wagon using machine vision. The system includes a prediction method for the operation time. Experimental results showed that the prediction error was 7.8%. The rate of work was 0.147ha/h in the prediction and 0.136ha/h in fact. To improve the accuracy of the prediction, it is necessary to measure the grain volume within the grain tank with higher precision, which will also improve the rate of work in the entire operation.
