Vision-Based Autonomous Vehicle Guidance for Indoor Security Patrolling by a SIFT-Based Vehicle-Localization Technique

摘要: A novel method for guidance of vision-based autonomous vehicles indoor security patrolling using scale-invariant feature transformation (SIFT) and vehicle localization techniques is proposed. Along-path objects to be monitored are used as landmarks localization. The work accomplished by three steps: SIFT-based object image matching, 2-D affine the Hough transform, analytic 3-D space transformation. Object monitoring can simultaneously achieved during vehicle-localization process, most planar-surfaced utilized in greatly enhancing applicability proposed method. Vehicle trajectory deviations from planned path due mechanic error accumulation also estimated setting up a calibration line on applying Moreover, path-correction technique conduct adjustment guide navigate next node. Analysis accuracy results finally included. experimental show that method, utilizing only single view each object, accurately monitor successfully.