Improving Efficiency in Agricultural UGVs Through Enhanced Pathfinding Techniques

Abstract

Precision agriculture leverages advanced technologies to increase the efficiency and automation of farming practices. Unmanned Ground Vehicles (UGVs) are central to this transformation, offering innovative solutions to automate and optimize agricultural tasks. This paper presents an innovative approach to optimize UGV robot navigation for obstacle avoidance in smart agriculture settings using an enhanced A-star algorithm. The primary navigation strategy integrates the Euclidean distance heuristic and Catmull-Rom splines for path smoothing. The A-star algorithm, traditionally effective in pathfinding due to its ability to combine actual travel cost and heuristic estimates, is further optimized to suit the complex terrain and obstacle dynamics typical of agricultural environments. Catmull-Rom splines generate smoother and more natural navigational paths, ensuring efficient and safe manoeuvring of the UGVs around obstacles, which is critical for crop monitoring and spraying. The methodology includes a comprehensive simulation environment where various obstacle configurations and terrain types are tested to validate the algorithm’s effectiveness. Results demonstrate that this hybrid approach significantly enhances path efficiency and safety, reducing operational time and improving the robot’s adaptability to new obstacles and changing terrain conditions, making it a viable solution for modern agricultural needs.