Energy Efficient UAV-Aided Vehicle Localization Frameworks

High precision seamless positioning is of great importance in various applications of autonomous vehicles,such as overtaking,intersection coordination,and platonooning,etc.Currently the vehicle location is mainly obtained from the Global Navigation Satellite Systems(GNSS)or Location Based Service(LBS)of base station.However,the positioning accuracy of GNSS can not meet the need of autonomous vehicles.And due to the existence of obstacles,GNSS signals may be attenuated,or even blocked in typical urban areas,where intelligent transportation systems are badly required.Nowadays,the unmanned ariel vehicle(UAV)is attractive due to its high mobility,large coverage and good line-of-sight propagations.The UAV is widely adopted in numerous applications such as search,rescue,communication relay and traffic control,etc.In this thesis,we try to use UAVs to assist vehicles in poor localization scenarios,to improve their location accuracy.However,the application of UAV is limited for its light weight and limited energy.Therefore,proper resource allocation plays an important role for vehicle localization and energy efficiency.We first derive the Cramer-Rao Lower Bound(CRLB)of the proposed vehicle localization framework,which can be used as the performance metrics.Both time of arrival(TOA)and a combination of time of arrival and angle of arrival(AOA)(TOA/AOA)measurements are adopted.Hybrid of TOA and AOA is attractive due to its advantages in limited anchors required.In the UAV-aided vehicle localization frameworks,the angle measurement information and the ranging information are obtained by the received signals,which is from antenna array.Then the vehicle position can be estimatede by fusing the two information,which also can improve the positioning accuracy of vehicle.Furthermore,the joint power and bandwidth allocation(JPBA)is performed among vehicles and UAVs,which can be relaxed as a standard semidefinite programming(SDP)problem,and soved efficiently.Since the power and bandwidth resources of the system(including both UAVs and vehicles)are usually limited,the JPBA strategy can improve the system resource efficiency and positioning accuracy of vehicle.In the vehicle positioning system,base on the TOA measurement or the TOA/AOA measurement,using the CRLB of the vehicle positioning accuracy as the optimization target,the JPBA strategy is adopted to reduce the vehicle positioning error.The simulation results show that the optimal allocation of vehicle positioning system resources can not only improve the vehicle positioning accuracy,but also improve the resource utilization rate of the system.Through the analysis of resource allocation scheme,it is found out that most of the resources are concentrated in UAV,and the links between UAVs to vehicles are not crucial.In addition,TOA/AOA measurement can improve the positioning accuracy of vehicles.