Ke, Tianqi (School of Vehicle and Mobility, Tsinghua University), Chen, Chaoyi (Tsinghua University), Xu, Qing (Tsinghua University), Cai, Mengchi (Tsinghua University), SUN, CHEN (Sony), Cui, Tao (Sony), Zhang, zhaoyu (Sony), Wang, Jianqiang (Tsinghua University), Li, Keqiang (Tsinghua University)

Estimation and Synchronization of Vehicle States with Uncertainty in Digital Twin System

Scheduled for presentation during the Invited Session "Driving the Edge: Addressing Corner Cases in Self-driving Vehicles" (WeBT5), Wednesday, September 25, 2024, 14:30−14:50, Salon 13

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on December 26, 2024

Abstract

Digital twins, a type of digital simulation technology, are essential for testing and deploying autonomous driving algorithms. Most of the existing systems rely on roadside cameras to capture vehicle states, resulting in uncertainty and suboptimal synchronization. In this paper, we model and analyze uncertainty factors related to connected roadside perception inputs, including perception uncertainty and communication uncertainty. The proposed methods for state estimation and synchronization incorporate communication corrections and an extended Kalman filter-based state estimation, supplemented by the Ackermann steering geometry algorithm for state synchronization. These approaches address uncertainty factors related to connected roadside perception inputs, enhancing system accuracy by reducing the disparity between perceived vehicle states and ground truth. Our methods effectively reduce the gap between perceived vehicle states and ground truth. Through simulation experiments, we validate the feasibility of our uncertainty-aware model for state estimation and synchronization.