Research on Target Approach Algorithm Based on GPS, Radar and Visual Image Fusion

With the rise in labor costs and the growing impact of human factors on errors, intelligent inspection equipment is gaining increasing popularity. However, most existing inspection systems for small vehicles primarily rely on GPS or radar navigation. Due to the inherent inaccuracies of GPS, significant positional errors often occur during airport positioning and navigation processes. This paper integrates GPS, vehicle-mounted 2D radar, and visual image recognition to design a target approach strategy algorithm. The proposed algorithm enables rapid and precise positioning. Experimental results demonstrate that, compared to systems relying solely on GPS, the small vehicle exhibits significantly improved accuracy in target approach and reduced heading angle deviation, achieving a target detection accuracy of 98.9%. Regardless of whether the target is positioned in front, behind, or to the sides of the vehicle, the system can automatically adjust and approach flexibly. These enhancements significantly boost the navigation performance and overall efficiency of the inspection vehicle.