Project Summary
The increasing deployment of CAVs presents both opportunities and challenges in navigating dynamic and complex environments such as highway work zones. Work zones introduce abrupt changes to road geometry, including lane closures, shifting traffic patterns, and the presence of workers and equipment, which pose significant challenges to CAV perception and planning systems. This research proposes a novel trajectory planning and optimization framework for CAVs operating in Smart Work Zones (SWZs), leveraging I2V communication for enhanced situational awareness. By integrating real-time data from smart vests, smart cones, and roadside units (RSUs), the system dynamically adjusts vehicle trajectories to ensure safety, efficiency, and compliance with work zone regulations.
The project will utilize real-world data collected in collaboration with Caltrans and validate the planning algorithms through both high-fidelity simulation and controlled field testing. A key component is the adaptation of the Enhanced Multi-Lane Planner from Baidu Apollo, optimized for the constrained and dynamic conditions of work zones. The planner incorporates infrastructure data, dynamic obstacle prediction, and vehicle kinematic constraints to generate smooth and safe trajectories. Ultimately, this research aims to bridge the gap between CAV autonomy and the complex nature of work zones, contributing to safer and smarter transportation infrastructure.
