Map Generation and Trajectory Analysis in Intelligent Transportation Systems

Map generation and trajectory analysis constitute two intertwined research pillars in intelligent transportation systems. Map generation encompasses the automated extraction and refinement of road networks from heterogeneous data sources such as vehicle GPS trajectories, remote sensing imagery and crowd-sourced navigation logs. Advanced methods employ geometric constructs like Delaunay triangulation, statistical models such as kernel density estimation and deep learning frameworks to detect road centrelines, intersections, lane configurations and dynamic attributes (for example, speed and traffic volume) in real time. Trajectory analysis focuses on interpreting the movement patterns of vehicles to infer traffic rules, turning behaviours, route preferences and network anomalies. Techniques range from clustering and sequence alignment algorithms to convolutional neural networks and spatio-temporal pattern mining. Together, these approaches enable the continuous update of high-precision maps, support adaptive route planning and underpin safety and efficiency enhancements in urban mobility. Global deployments have demonstrated practical applications in navigation services, traffic management and infrastructure planning, highlighting the vital role of data fusion and machine intelligence in modern transport ecosystems.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Map Generation and Trajectory Analysis in Intelligent Transportation Systems publication trend

The graph below shows the total number of articles in map generation and trajectory analysis in intelligent transportation systems across all publications each year (not limited to Nature Index journals).

Technical terms

GPS trajectory: A sequence of time-stamped geospatial positions recorded by a moving vehicle or device.

Convolutional neural network (CNN): A deep learning architecture designed to extract hierarchical spatial features, commonly used for image analysis and pattern recognition.

Delaunay triangulation: A method of connecting a set of points to form triangles such that no point lies inside the circumcircle of any triangle, useful in boundary and network reconstruction.

Dynamic time warping (DTW): An algorithm for aligning two sequences that may vary in time or speed, often applied to compare movement trajectories.

Kernel density estimation (KDE): A non-parametric statistical technique for estimating the probability density function of a random variable, used in spatial analysis to detect clusters such as road intersections.

References

1. A Road Map Refinement Method Using Delaunay Triangulation for Big Trace Data. ISPRS International Journal of Geo-Information (2017).
2. Fusing Taxi Trajectories and RS Images to Build Road Map via DCNN. IEEE Access (2019).
3. Trajectory analysis at intersections for traffic rule identification. Geo-spatial Information Science (2020).
4. A Hybrid Method to Incrementally Extract Road Networks Using Spatio-Temporal Trajectory Data. ISPRS International Journal of Geo-Information (2020).