Vehicle Localization utilizing a Novel Hybrid TDOA-Based Estimation

Abstract

This research investigates the use of a hybrid technique to locate vehicle positions on a 2D plane solely via other vehicles to further the future realization of Vehicle-to-Vehicle (V2V) communication. An approach in which trilateration and Time Difference Of Arrival (TDOA) are combined to estimate the Direction Of Arrival (DOA) of an incoming signal is considered. By using TDOA measurements of receivers on the Receiver Vehicle (RV), estimation regions are constructed to robustly obtain the Transmitter Vehicle (TV) position. This proposal not only creates a method for TDOA to be directly used in V2V communication but compared to other localization methods such as TOA (Time Of Arrival), the proposed technique does not need to consider time synchronization between the TV and RV, allowing for usage in a larger variety of on-road scenarios. A regression model is also implemented to further improve the accuracy of the estimation. Evaluation of the proposal is conducted for same side DOA and opposing side DOA. The DOA estimation was compared with a theoretically ideal scenario incorporating TOA. For further clarification of the methods utility and to mimic the transmission signal in road environments, the proposal is also tested in a ray tracing propagation model. The simulations show that the proposed solution accompanied with the regression model estimated the DOA in a 1 nanosecond (ns) time step environment to 1.92° accuracy and 0.08°accuracy in a 0.1ns time step environment.