Modeling an Intelligent Framework for Optimizing UAV Path Planning and Anti-collision in Agriculture

Abstract

Unmanned aerial vehicles (UAV) are increasingly utilized for monitor expansive farming to their effectiveness in observation and data gathered. Efficient path planning and collision prevention are critical for optimizing UAV operation in such settings. The efficiency can be controlled by the quality and declaration of UAV sensed data, as well as the difficulty of real world ecological variables that could impact path optimization and collision prevention. This research introduce a novel technique, the customizable dung beetle search tuned random forest (CDBS-RF) method, designed to improve UAV route planning. The CDBS-RF method integrated the dung beetle search (DBS) algorithm, known for its robust optimization capabilities, with random forest (RF) to enhance optimal path security and effectiveness. This method dynamically adjusts path safety and effectiveness. This approach dynamically adjusts path planning parameter to make sure optimal route selection and collision prevention. The suggested technique was evaluated utilizing UAV-sensed data and implement in python based virtual environment. Experimental consequences demonstrate that the CDBS-RF approach considerably enhances UAV path planning performance, provide safer and more efficient navigation for self and more efficient navigation for self-sufficient tarp monitoring. The performance evaluation techniques include the planning time (0.789) and path length (21.526). By utilize advanced optimization and anti-collision algorithms, this technique offer a promising solution for improving UAV operations in agricultural surveillance.