Distance-Based Cooperative Localization of Connected Vehicles Via Convex Relaxation Under Extreme Environments

This paper proposes a distance-based measurement semidefinite programming (SDP) connected and automated vehicles (CAVs) cooperative localization method under extreme environments. The main works of this paper are threefold. First, considering that communications link is nonstraight linear and its propagation distribution may be unknown, a novel cooperative localization algorithm based on SDP is proposed to reduce propagation in severe urban forest environments. Additionally, the mean square error (MSE) of target vehicle localization under different dimensions and different variances are considered. Finally, the effectiveness of the proposed algorithm in existing methods under different conditions is compared. Simulation comparison results indicate that the proposed convex relaxation method is better than the existing methods and achieves a performance closer to Cramer-Rao lower bound (CRLB).