Dynamic Goal-Based Adaptive Line of Sight for 3D Path Following: An X300 Case Study

This paper presents a Dynamic Goal-Based Adaptive Line-of-Sight algorithm for three-dimensional path-following in autonomous vehicles, addressing limitations in traditional Line-of-Sight methods that use fixed lookahead distances. The proposed approach adjusts the lookahead distance dynamically based on real-time feedback from cross-track errors, vertical-track errors, and path curvature, improving stability and accuracy in complex environments. Five simulation trials using the X300 Autonomous Underwater Vehicle were conducted, varying both lookahead distance (∆) and velocity(v). Results show that the dynamic method significantly reduces cross-track and vertical-track errors, outperforming fixed lookahead approaches, especially in sharp turns and challenging conditions.