O'Keefe, KyleWang, Da2015-02-252015-06-232015-02-252015http://hdl.handle.net/11023/2109Many intelligent transportation systems applications require precise relative vehicle position. Global Navigation Satellite Systems, particularly GPS currently provide this through either absolute or differential positioning. GPS performance is limited in environments with degraded or block signals. This thesis proposes to augment differential GPS (DGPS) with range and bearing observations to surrounding vehicles and infrastructure and accomplishes this by tightly integrating DGPS, range and bearing observations in a small network of vehicles or infrastructure points. The performance of this system is assessed using real GPS, and Ultra-Wide Band (UWB) ranging radio observations and “simulated” bearing data. The integrated solution outperforms the DGPS only solution. A Vehicle-to-Infrastructure (V2I) test at a deep urban canyon intersection show sub-metre to metre level horizontal positioning accuracy with three UWB ranging radios deployed at intersection, compared to tens of metres accuracy of DGPS only. For Vehicle-to-Vehicle (V2V), the DGPS and UWB outperforms DGPS only by 10%. Systematic UWB range errors are effectively estimated if integrated with the DGPS carrier phase Real Time Kinematic (RTK) solution. As a result, the UWB ranges improves the convergence of the carrier phase RTK float solution and the time to fix ambiguities. A full-order decentralized filter with post estimation information fusion was developed for V2V cooperative navigation. The full-order decentralized estimate on each vehicle can be fused with the estimate of other vehicles to achieve the centralized equivalent estimate, even if these estimates have different nuisance error states (including UWB systematic errors and carrier phase ambiguities), by fully taking account of the correlation in the observations and state covariance the filter has, which is demonstrated using GPS data and UWB range data collected in three-vehicle V2V field tests. In other cases if some of the correlation between the nuisance error states and the position states is being ignored, the vehicle that has access to fewer observations can still also benefit from the cooperation via fusing its estimate with that of another vehicle that has better solution.engUniversity of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission.RoboticsDGPSUWB rangeBearingV2X positioningDecentralized filtering and fusiondoctoral thesis10.11575/PRISM/25464