Grid and Graph-based Autonomous Transportation System State Awareness Open Dataset

1. Dataset Overview This dataset is based on 7 days of connected vehicle trajectory data from a portion of the road network in Jiading District, providing an open dataset for autonomous transportation system state awareness. The dataset includes traffic flow, average speed, and other metrics, which can be used for traffic flow prediction, traffic pattern recognition, path optimization, spatiotemporal analysis, and other related research. 2. Dataset Content This dataset contains spatiotemporal traffic state awareness data based on graph and grid models for a specific area in Jiading, covering the period from April 15 to April 21, 2024 (7 days in total). The road network data for Jiading is obtained from OpenStreetMap, and road segment IDs are assigned based on this data. The individual vehicle origin-destination (OD) and trajectory data are collected in a set of trajectory files with a sampling frequency of 1Hz. Each trajectory file contains timestamped data for each frame, including vehicle ID, latitude, longitude, and corresponding road segment ID (the sequence of road segment IDs forms the path). Notably, for privacy protection, each vehicle is assigned a unique vehicle ID for each trip, and each vehicle ID corresponds to only one trip's OD and path sequence. 3. Data Files The dataset includes the file `jiading_data.zip`, which contains two folders: ShanghaiGraph and ShanghaiGrid, storing graph-based and grid-based traffic data, respectively. 4. Data File Description The file structure of the dataset is as follows: jiading_data.zip ├── ShanghaiGraph/ │ ├── graph_data.npz # Graph data │ └── ShanghaiGraph.csv # Spatial adjacency relationship of the graph structure, with the "cost" field representing the actual distance (m) └── ShanghaiGrid/ ├── grid_data.npz # Grid data └── ShanghaiGrid.csv # Spatial adjacency relationship of the grid structure, with the "cost" field representing the actual distance (m) The contents of `graph_data.npz` include the following fields:['data', 'data_nxtc', 'time_slots', 'node_count', 'time_steps', 'feature_names'] | Column Number | Type | Description ||---------------|-----------------|-----------------------------------------------------|| 1 | data_nxtc | Data for the next time step, typically used for time series prediction tasks || 2 | data | Graph data for the current time step, typically represented as an adjacency matrix or other data related to node relationships || 3 | time_slots | Time slot information, indicating the specific time of each data point || 4 | node_count | Number of nodes in the graph || 5 | time_steps | Number of time steps, representing the time dimension of the data || 7 | feature_names | Feature names, listing all the features in the dataset and what each column represents | The contents of `grid_data.npz` include the following fields:['data_nxtc', 'data', 'dates', 'node_count', 'time_steps', 'day_count', 'feature_names'] | Column Number | Type | Description ||---------------|-----------------|-----------------------------------------------------|| 1 | data_nxtc | Data for the next time step, typically used for time series prediction tasks || 2 | data | Grid data for the current time step, typically represented as an adjacency matrix or other data related to node relationships || 3 | dates | Date information, representing the specific date of each data point || 4 | node_count | Number of nodes in the grid || 5 | time_steps | Number of time steps, representing the time dimension of the data || 6 | day_count | Number of days in the dataset, representing the range of dates || 7 | feature_names | Feature names, listing all the node features in the grid data | 5. Data Statistics Graph Dataset:- Number of nodes: 2752- Average missing rate: 42.22%- Time slot length: 30 minutes- Traffic flow statistics: - Mean: 66.86 - Median (p50): 18 - 95th percentile (p95): 254 - Max: 5239 (non-zero positions)- Average speed statistics: - Mean: 38.99 - Median (p50): 30.75 - 95th percentile (p95): 90.67 - Max: 125.1 (non-zero positions)- Average traffic flow per node per time slot: 38.63 (including missing data)- Current algorithm accuracy: - MAE: 7.06 - RMSE: 11.13 - MAPE: 52.26% Grid Dataset:- Grid size: 1km × 1km- Number of grids: 624- Average missing rate: 47.9%- Time slot length: 30 minutes- Traffic flow statistics: - Mean: 121.29 - Median (p50): 37 - 95th percentile (p95): 541 - Max: 2597 (non-zero positions)- Average speed statistics: - Mean: 37.27 - Median (p50): 31.12 - 95th percentile (p95): 83.38 - Max: 127 (non-zero positions)- Average traffic flow per grid per time slot: 63.1971- Current algorithm accuracy: - MAE: 6.31 - RMSE: 9.42 - MAPE: 72.80% 6. Data Provider This dataset is provided by Beijing University of Posts and Telecommunications, funded by the National Key Research and Development Program of China under the project "Computing Technologies for Autonomous Transportation Systems (2023YFB4301900)".