Calibration and Georeferencing for GNSS-Equipped Vehicle Video Mapping Using a Tesla Model Y

Abstract

Abstract. The evolution of mapping platforms has followed a consistent pattern: professional instruments are complemented by consumer devices that trade precision for scalability. Unmanned aerial systems transformed aerial photogrammetry by making it accessible beyond traditional aircraft, and smartphones, especially when equipped with high accuracy GNSS positioning, have demonstrated viable terrestrial mapping. This paper extends that progression to vehicle-based mapping by presenting SurveyXR, a web-based calibration and georeferencing framework that converts dashcam video from a GNSS-equipped vehicle into georeferenced imagery suitable for Structure-from-Motion (SfM) processing. By providing accurate per-frame exterior orientation parameters, the system enables direct georeferencing of the SfM output, eliminating the need for ground control points in the photogrammetric workflow. The pipeline implements checkerboard-based intrinsic calibration with automated quality diagnostics, Perspective-n-Point exterior orientation solving with automatic boresight detection, GNSS-synchronized frame extraction, and lever arm correction between the GNSS antenna and each camera. All computation runs in a browser or lightweight cloud backend, requiring no local software installation. The framework was evaluated on a 2026 Tesla Model Y equipped with a roof-mounted Emlid Reach RS4 Pro PPK GNSS receiver on the Ohio State University campus. Georeferencing accuracy was assessed against 71 independently surveyed RTK check points in two configurations: direct georeferencing only (no ground control) and GCP-constrained bundle adjustment. The paper documents the calibration methodology, time synchronization model, error budget analysis, and quantitative accuracy assessment.