Download files
Abstract
Road transportation injuries, environmental pollution, and wasted energy and time in traffic congestion cause considerable cost to society. Some examples are the $115 billion cost of United States traffic congestion in 2009 and its predicted $23 billion annual cost in Australia by 2020. Intelligent Transportation Systems (ITS) are increasingly being considered to mitigate these impacts. The benefit-cost ratio of 9-1 for ITS has led to significant investment in this industry. Therefore, there is an increased need for research in fields such as vehicular positioning, a fundamental part of many ITS applications. Although Global Navigation Satellite Systems (GNSS) are applicable for navigation and fleet management, the accuracy does not meet the requirements for the systems such as collision avoidance and lane-level positioning. Limited availability of GNSS in urban canyons is another challenge for positioning. These issues create a gap between the positioning performance required for ITS and that which GNSS provides. Bridging this gap, this research considers Cooperative Positioning (CP) based on vehicular communication. The applicability of vehicular CP techniques proposed in the literature is questionable due to different viability issues. Focusing on viability, techniques are proposed here for different situations in terms of the availability of GNSS, specifically the Global Positioning System (GPS). Dedicated Short Range Communication (DSRC), the standard medium for vehicular communication, is considered for CP purposes.
CP techniques proposed in the literature tend to rely on range estimates among participating nodes. The infeasibility of this approach in vehicular environment is demonstrated. Range-rate-based CP is introduced as an alternative using DSRC Carrier Frequency Offset (CFO). Typical achievable performance for different CP approaches is investigated. Range-rate-based CP techniques and new methods without ranging or range-rating are proposed. For full GPS coverage, two CP techniques, loose and tight integration, are introduced which can improve GPS-based accuracy and precision by up to 47%. For low numbers of visible satellites, a CP method is proposed for 3D positioning using two GPS satellites. A CFO-based CP method is presented for the situations without GPS coverage. This technique provides an instantaneous lane-level positioning which can improve the performance and functionality of many ITS applications.