Cycle-Level Distribution-Based Calibration of Microscopic Intersection Models Using UAV Trajectories
summary
A UAV-driven SUMO calibration pipeline is built that targets signal cycle-level cumulative exit-count curves instead of link totals, reinitializes each cycle from UAV-derived states so cycles run independently and in parallel, and jointly tunes Krauss car‑following, sublane lane‑changing and junction interaction parameters. Optimization uses the NGOpt optimizer from Nevergrad and yields stable convergence with a dominant-movement nABC as low as 0.46% and at most two vehicles per cycle deviation, reducing median cycle MAPE by ~70% on high‑volume movements.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:49Why intersection calibration matters
- 2:53UAV trajectories for calibration
- 3:55Calibration framework and inputs
- 5:03Trajectory processing and simulation
- 7:12Cycle-based objective evaluation
- 8:35Athens intersection setup
- 10:04Signal timings from UAV data
- 11:30Parameter sensitivity screening
- 12:56Mixed traffic in SUMO
- 13:59Preliminary results and metrics
- 15:23Optimization convergence and validation
- 16:30Discharge curve comparisons
- 17:34Key findings and contributions
- 18:14Future research and conclusions
related talks
16mAutomated Calibration of Traffic Demand and Traffic Lights in SUMO Using Real-World Observations
13mUtilizing Vehicle Trajectory Data to Characterize Performance Measures of a Highway Corridors
16mOnline calibration with SUMO for network-wide traffic and emission monitoring - Case ITS Huainan
16mCombining microtraffic models and complex vehicular simulators
13mExtending SUMO for Lane-Free Microscopic Simulation of Connected and Automated Vehicles
20mSimulating platooned connected autonomous vehicle in a large scale urban scenario
17mSignal priority for improving fluidity and decreasing fuel consumption
17mA comparison of SUMO’s count based and countless demand generation tools