80# Awesome Trajectory Datasets 
A curated list of open datasets containing road traffic trajectories (vehicles and vulnerable road users), with a focus on large-scale, real-world data in global coordinate formats, suitable for traffic analysis, motion prediction, and autonomous systems research.
Several datasets below are already integrated and publicly explorable on Trajectory Trace. Look for the 🔍 badge.
 |
CitySim - A Drone-Based Vehicle Trajectory Dataset for Safety Oriented Research and Digital Twins (2023): 12 locations (USA) · drone · ~1,140 trajectories, 30 Hz· CC-BY-NC 4.0 · Data on Request · Paper |
 |
DLR UT - DLR Urban Traffic (2023): Brunswick (Germany) · stereo cams · 12 h, 20 Hz, inner ring road · trajectories, traffic lights, local weather, air quality, road conditions · CC-BY 4.0 · Direct Download · Paper · 🔍 |
|
DrivIng - A Large-Scale Multimodal Driving Dataset with Full Digital Twin Integration (2025): Ingolstadt (Germany) · 10 Hz cams, lidars, 18km ego vehicle · CC BY-NC-ND 4.0 · Direct Download · Paper |
|
GeoLife - Geolife GPS trajectory dataset (2012): · Beijing (China) · GNSS · 182 users, 5 years, multi-modal, 0.2-1 Hz · MSR-LA (NC) · Direct Download · User Guide |
|
INTERACTION - INTERnational, Adversarial and Cooperative moTION Dataset (2019): Multi-country · 10 Hz drone or cam · ~16 locations: merges, roundabouts, intersections · No explicit license provided by the authors · Data On Request · Paper · GIT |
 |
OpenDD - A Large-Scale Roundabout Drone Dataset (2020): 7 roundabouts (Germany) · 84k+ tracks, over 62 hours · CC-BY-ND 4.0 · Direct Download · Paper · 🔍 |
|
SinD - A Drone Dataset at Signalized Intersection in China (2022): Changsha (China) · drone · ~14k tracks at signalized intersections · CC0 1.0 Universal · Demo data in Repo, Full Data on Request · Paper |
 |
TUMDOT-ING - Trajectories from Urban Multimodal Drone Observations of Traffic Ingolstadt (2023): Ingolstadt (Germany) · 6 drones, 30 Hz · 3 locations, 3 days, ~3 h each · covers the high definition testfield Ingolstadt (see UrbanIng-V2X) · CC-BY-NC 4.0 · Direct Download · 🔍 |
 |
TUMDOT-MUC - Trajectories from Urban Multimodal Drone Observations of Traffic Munich (2022): Munich (Germany) · 12 drones · 6 locations, >3 h each, 700 m continuous section along Rheinstraße, 25 Hz · CC-BY-NC 4.0 · Direct Download · Paper · 🔍 |
 |
UrbanIng-V2X - A Large-Scale Multi-Vehicle, Multi-Infrastructure Dataset Across Multiple Intersections for Cooperative Perception (2025): Ingolstadt (Germany) · LiDAR, rgb cam, thermal cam, GNSS/IMU · V2X cooperative perception at 3 intersections, 34 sequences × 20 s, ~712k annotated instances at 10 Hz · CC-BY-NC-ND 4.0 · Direct Download · GIT · Paper · 🔍 |
 |
Pedestrian Trajectory Dataset of Public European Squares - Includes 348,300 pedestrian trajectories extracted from 193 hours of video footage, collected at morning, noon, evening and saturdays (2024): 39 city squares (Europe) · raw data at 15 Hz, processed data at 2 Hz · CC-BY 4.0 · Direct Download · Paper · GIT · 🔍 |
 |
Automatum Data: Bavaria (Germany) · drone · 30 h · CC BY-ND / demo · Data On Request · Paper · 🔍highway · 🔍roundabout |
 |
DLR HT - DLR Highway Traffic (2024): Brunswick (Germany) · Test Bed Lower Saxony · 20 Hz trajectories, weather, road condition · CC-BY-NC-SA 4.0 · Direct Download · Paper · 🔍 |
 |
MiTra - Milan Trajectories (2024): Milan (Italy) · 900 m section of the A50 urban freeway · 135 min, using 6 drones, 30 Hz · CC BY 4.0 · Direct Download · Paper · GIT · 🔍 |
 |
NGSIM - Next Generation Simulation Vehicle Trajectories and Supporting Data (2006): California (USA) · roadside cam · ~9,000 tracks, 10 Hz · CC BY-SA 3.0 · Direct Download · Docs · 🔍 |
 |
pNEUMA - Open large-scale dataset of naturalistic trajectories of half a million vehicles (2018): Athens (Greece) · 10 drones, 25 Hz · 4 days, each ~4 h, city center · CC-BY-NC 4.0 · Direct Download · Paper · 🔍 |
|
SWIFTraj - Swarm Intelligence Freeway–Urban Trajectories Dataset (2022): Nanjing (China) · 16 drones, 10 Hz · 4.5km Hurong Expressway, 4 hours · CC BY 4.0 · Data on Request · Paper · GIT |
Contributions are very welcome! Please read the contributing guidelines before submitting a PR.
Quick checklist for adding a dataset or live feed:
- The data is publicly accessible (free download, registration, or free-tier API)
- It contains trajectory data (x/y/t - positions over time)
- A citable source (paper, DOI, or official website) exists
- The license is clearly stated
- For live feeds: the protocol and update rate are documented
If your dataset shall be integrated into Trajectory Trace, mention it in the PR - we'll integrate it and add the 🔍 badge.
This Source Code Form is subject to the terms of the Apache License 2.0. If a copy of the APL2 was not distributed with this file, You can obtain one at https://www.apache.org/licenses/LICENSE-2.0.txt.