---
license: cc-by-nc-4.0
pretty_name: 15 Laps of 30Hz NGSIM-Style Telemetry from UE5 Physics Simulation
task_categories:
- tabular-classification
- reinforcement-learning
- robotics
language:
- en
tags:
- UE5
- Autonomous-Driving
- Telemetry
- Human-in-the-loop
- Chaos-Physics
- NGSIM
size_categories:
- 1K<n<10K
---

# Dataset Card for 15 Laps of 30Hz NGSIM-Style Telemetry

This is a Lite Version of a larger research dataset focusing on human driving signatures in high-fidelity simulations. It includes 15 full laps of telemetry captured at 30Hz within Unreal Engine 5, specifically formatted to match NGSIM standards.

## Dataset Details

### Dataset Description

This Lite Version dataset contains 15 laps of high-fidelity human driving telemetry. It is intended for researchers and students to verify data compatibility and test behavioral cloning scripts using a 27-dimensional input vector (Steering, Throttle, 12-Ray Sensors, etc.).

* **Curated by:** Atlas Built
* **Language(s) (NLP):** English
* **License:** cc-by-nc-4.0

### Dataset Sources

* **Repository:** [https://atlasbuiltit.etsy.com/listing/4500776358](https://atlasbuiltit.etsy.com/listing/4500776358)

## Uses

### Direct Use

Verification of data compatibility for vehicle simulation pipelines and testing of initial behavioral cloning or reinforcement learning scripts.

### Out-of-Scope Use

Commercial use without explicit permission; safety-critical real-world autonomous driving deployment without further validation.

## Dataset Structure

The data is structured in .CSV format following NGSIM-style telemetry standards. It includes vehicle control inputs (Throttle, Brake, Steering), kinematic data (Velocity, Heading), and environmental perception data (12-Ray sensor arrays).

## Dataset Creation

### Curation Rationale

To provide a lightweight, accessible sample of high-fidelity simulation data for researchers focusing on human-in-the-loop driving and shared control.

### Source Data

#### Data Collection and Processing

Data was collected at a constant 30Hz sampling rate using Unreal Engine 5 Chaos Physics vehicle templates during manual human-driven sessions.

#### Who are the source data producers?

Human operators interacting with the UE5 vehicle physics simulation.

#### Personal and Sensitive Information

The dataset contains no personal or sensitive information. It consists entirely of mechanical and environmental telemetry from a simulated environment.

## Bias, Risks, and Limitations

The dataset is limited to a specific vehicle physics model in UE5 and may not perfectly represent real-world tire-road dynamics. It represents a specific subset of driving behaviors (15 laps).

### Recommendations

Users should be made aware of the risks, biases and limitations of the dataset. This is a Lite Version; for larger-scale training, the full 120,000+ row dataset is recommended.

## Dataset Card Contact

[https://atlasbuiltit.etsy.com/](https://atlasbuiltit.etsy.com)