Traffic Scanner (TScan)

TScan is a LiDAR-based traffic observation system developed by Purdue University's Center for Road Safety to capture detailed movements of vehicles, pedestrians, and bicyclists at intersections and roadway sites. By converting three-dimensional sensor measurements into road-user trajectories, TScan supports multiple traffic-engineering and safety applications from a common source of movement data.

The research program has progressed from stationary object detection and tracking, to field-deployable trailer systems, and then to proactive safety analysis based on encounters and conflicts.

Why LiDAR?

LiDAR measures the surrounding environment directly in three dimensions. Unlike conventional video, which represents a three-dimensional scene as a two-dimensional image, LiDAR provides spatial measurements that can be used to detect road users and reconstruct their trajectories. These capabilities support TScan's goal of capturing detailed traffic movements at intersections.

Core Capabilities

TScan carries field measurements through a connected sensing and analysis process.

01

Road-User Detection

Separating the stationary environment from moving vehicles, pedestrians, and bicyclists.

02

Classification & Tracking

Classifying road users and maintaining continuous tracks as they move through the monitored area.

03

Trajectory-Based Measurement

Deriving paths, speeds, movements, and spatial patterns from detailed road-user trajectories.

04

Proactive Safety Analysis

Using encounters and conflicts to examine safety conditions during comparatively short field observations.

From Research Concept to Field-Deployable System

TScan evolved through a sequence of Center projects sponsored through JTRP by INDOT.

01

Detecting and Tracking Road Users

The initial research developed signal-processing methods to remove the stationary background, detect and track moving objects, classify road users, and store their trajectories for engineering analysis. A LiDAR scanner was integrated with Purdue's Mobile Traffic Laboratory to demonstrate feasibility.

View SPR-3831 report
02

Moving Toward Real-World Implementation

Two trailer-based prototypes were constructed, and the detection and tracking algorithms were converted into a C++ implementation. Multiple sensors were brought into common spatial and temporal reference frames, supporting traffic counting, speed analysis, and pedestrian-presence visualization.

View SPR-4102 report
03

From Trajectories to Proactive Safety

Subsequent research applied TScan trajectories to traffic encounters and conflicts. The work supported visualization of interaction patterns, short-period crash estimation, and expansion of conflict-based estimates toward annual values.

View SPR-4439 report
Mobile Traffic Laboratory, retractable mast-mounted LiDAR and cameras, and a LiDAR point-cloud visualization
Clockwise from top left: Mobile Traffic Laboratory with telescopic mast extended; mast-mounted Velodyne LiDAR and Vivotek PTZ cameras; and an example LiDAR point-cloud visualization.

What Can TScan Measure?

The trajectory-based approach moves from aggregate observations toward detailed road-user movement data.

  • Road-user detection, classification, and tracking
  • Vehicle, bicycle, and pedestrian trajectories
  • Traffic movement counts
  • Vehicle speed analysis
  • Pedestrian-presence visualization
  • Traffic encounter and conflict analysis
  • Spatial visualization of traffic interactions
  • Short-period safety and crash estimation

The Common Thread: Trajectories

At the heart of TScan is the conversion of LiDAR measurements into road-user trajectories. Rather than reducing an intersection immediately to aggregate statistics, those trajectories preserve detailed information about how individual road users move through the monitored area. Different engineering applications can then use the same underlying movement data for different analyses.

Reports and Suggested Citations

The three reports document the progression from sensing and tracking to field implementation and proactive safety analysis.

SPR-3831 · 2016

Tarko, A. P., Ariyur, K. B., Romero, M. A., Bandaru, V. K., & Lizarazo, C. G. (2016). TScan: Stationary LiDAR for traffic and safety studies—Object detection and tracking (Joint Transportation Research Program Publication No. FHWA/IN/JTRP-2016/24). Purdue University.

View report
SPR-4102 · 2021

Tarko, A. P., Romero, M. A., Ariyur, K. B., Bandaru, V. K., & Lizarazo, C. G. (2021). TScan–Stationary LiDAR for traffic and safety applications: Vehicle interpretation and tracking. Purdue University.

View report
SPR-4439 · 2023

Tarko, A. P., Bandaru, V. K., & Romero, M. A. (2023). Guidelines for evaluating safety using traffic encounters: Proactive crash estimation on roadways with conventional and autonomous vehicle scenarios. Purdue University.

View report