Accurate localization in canopy-occluded, GNSS-challenged environments is critical for autonomous robots and intelligent vehicles. This paper presents a coarse-to-fine trajectory estimation framework using millimeter-wave radar as the primary sensor, leveraging its foliage penetration and robustness to low visibility. The framework integrates short- and long-term temporal feature enhancement to improve descriptor distinctiveness and suppress false loop closures, together with adaptive OpenStreetMap-derived priors that provide complementary global corrections in scenarios with sparse revisits. All constraints are jointly optimized within an outlier-robust backend to ensure global trajectory consistency under severe GNSS signal degradation. Evaluations conducted on the MulRan dataset, the OORD forest canopy dataset, and real-world campus experiments with partial and dense canopy coverage demonstrate up to 55.23% reduction in Absolute Trajectory Error (ATE) and a minimum error of 1.83 m compared with baseline radar- and LiDAR-based SLAM systems. The results indicate that the integration of temporally enhanced radar features with adaptive map constraints substantially improves large-scale localization robustness.

Continuous stability, as one of the core modules of the autopilot system, is particularly important for its performance. However, as the vehicle speed increases, the system positioning error may be amplified, consequently introducing deviations in the positioning consistency of the system. The inherent high speeds and motion constraints in highway environments introduce new challenges for feature matching, particularly in vision-based vehicle localization, where initialization and scale estimation biases are further expanded. Lane markings, characterized by their simple and uniform structures and high distinctiveness from the surrounding environment, serve as effective features for matching-based localization in autonomous driving. This letter introduces a high-precision and robust vehicle localization method based on lane model constraints. Initially, leveraging lane model parameters from prior maps, we track and model lane line detections across consecutive frames to enhance the completeness of lane representation. The tracking results, combined with prior map data on lane widths, are employed to optimize scale parameters. Subsequently, real-time detected lanes are matched with prior maps through point-map association to constrain the vehicle’s heading angle. Finally, map matching results are integrated into existing visual local odometry methods to perform real-time localization optimization, thereby improving localization performance. Experimental evaluations conducted on a self-collected highway dataset demonstrate that the incorporation of lane models significantly enhances system localization accuracy.