Radar-Based Evaluation of Car-Following Behavior and Fuel Consumption Across Vehicle Categories
summary
Radar trajectories from a 1.5 km signalized corridor with 1,815 follower trips (250 cars, 607 SUVs, 878 light‑duty trucks, 80 Class‑8) are replayed in SUMO via TraCI while followers run the IDM, comparing default and trajectory‑level calibrated parameters. Calibration reduces RMSE in spacing, velocity, acceleration for all categories, spacing RMSE for Class‑8 falls from 14.95 to 3.48, but Class‑8 still shows the largest residual error. Fuel estimates from PHEMlight decline after calibration, median fuel drops 12.7% for Class‑8, and using a passenger vs trailer vClass for Class‑8 inflates uncalibrated errors until calibration corrects them.
AI-written summary · about