DBRow: A Density-Based algorithm for autonomous navigation within crop rows

This paper introduces DBRow, a density-based algorithm designed to improve autonomous navigation within crop rows, addressing the growing need for efficient agricultural robotics to boost productivity and tackle labour shortages.DBRow integrates Simultaneous Localisation and Mapping (SLAM) with Density-Based Spatial Clustering of Applications with Noise (DBSCAN), overcoming the limitations of previous navigation systems that relied solely on LIDAR data for NMBU's FRE participation.Experiments conducted in simulated and controlled indoor environments evaluated DBRow using A* path planning algorithm.The results show some weaknesses in the simulated environment, but it performs well in the controlled indoor environment.The paper calls for further testing for statistically significant results and suggests future enhancements, including LIDAR preprocessing improvements and machine learning integration, to optimise navigation accuracy and automate tasks like pesticide application.