Editorial: Intelligent Vehicle Navigation (part 2)

This is the second of two parts of this Special Issue on Intelligent Vehicle Navigation (iVN). The use of in-vehicle navigation systems continues to rise globally.. The capability of these systems in terms of positioning accuracy and navigational services and their sophistication in terms of system design and reliability have also advanced over time. This has been possible due to iVN supported by enhanced navigation sensor and digital spatial road network data and intelligent navigation algorithms. Conventional methods for vehicle navigation have employed GPS as the main tool integrated with Dead Reckoning (DR) sensors (in various configurations) and 2D spatial (map) data using map matching algorithms. Recently, other techniques that employ technologies of opportunity including wireless positioning, 3D spatial (e.g. city) models together with vision systems (cameras and videos) have emerged, to improve navigation performance in areas where GPS is unable to deliver the required performance. The challenge here is how to integrate the various systems, sensors and databases to achieve optimal navigation performance. This second part of the special Issue of the Journal of Transportation Systems on iVN continues the theme of the first part and presents research that addresses aspects of the problem of vehicle navigation including integration of GPS with other sensors and databases for improved vehicle navigation. the feasibility of integrating GPS with 3D spatial (city) model and vision systems (camera) for vehicle navigation. Instead of using data from the common DR sensors of vehicle odometer and a gyroscope for positioning during periods when GPS only navigation is not possible, the proposed system determines the vehicle position by DR (distance and direction data) from the 3D spatial model calibrated by real images from an on-board camera. Tests using real-world data collected in Nancy in France show that the performance offered by 3D spatial model/camera system in terms of positioning accuracy is similar (i.e., average error 4m) to that of the GPS calibrated odometer/gyroscope system during GPS outages. Further research is required to characterise fully the performance in different operational environments. The second paper, " Methods to detect road features for video-based car navigation systems " framework to detect road features in different s road conditions using advanced methods including Support Vector Machines (SVMs) and Bayesian network models. The detection such road features is fundamental to support sophisticated navigation services such as providing directional information to the drivers and assisting them to change lane at the appropriate …