This study introduces a framework for controlling large-scale mixed traffic using a privacy-protecting crowdsensing method, a graph propagation-based traffic forecasting method, and a privacy-preserving route selection mechanism. The framework was evaluated on a real-world road network, showing accurate traffic flow forecasts, efficient mitigation of network-wide robot vehicle shortage, and effective coordination of large-scale mixed traffic. Compared to other methods, the framework reduced the robot vehicle shortage issue by up to 69.4% and the average waiting time by up to 27%.
Publication date: 21 Nov 2023
Project Page: unknown
Paper: https://arxiv.org/pdf/2311.11347
