Lane-Relative Pose Localization Dataset
- Records
- 4560
- Size
- 0.4GB
- Format
- JSON
- Provider
MOBIUSI INC- Updated
- v1.02026-11-04
This dataset uses vehicle pose, lane geometry, and coordinate references as inputs to provide structured localization results relative to lane centerlines and lane segments. Outputs include the nearest centerline point index, longitudinal position, lateral offset, heading difference, projected point, lane-segment progress, and lane-membership status, with explicit units and value conventions. It supports structured-output supervised fine-tuning, lane-level localization development, map matching, and spatial relationship evaluation; sample volume and coverage depend on the specific data offering.
Overview
Dataset Insights
Preview
Sample Examples
Schema
Technical Specifications
| Field | Type | Description |
|---|---|---|
| file_name | string | File name |
| scene_input | object | Raw JSON input containing vehicle pose, lane geometry, and coordinate references. |
| is_within_lane | boolean | Indicates whether the vehicle is within the valid lane area, based on the absolute lateral offset and lane half-width. |
| lateral_offset | number | Signed lateral distance in meters from the vehicle reference point to the lane centerline. The sign follows the lane travel direction and the right-hand coordinate convention. |
| heading_difference | number | Normalized angular difference in radians between the vehicle heading and the tangent direction of the lane centerline, ranging from negative pi to pi. |
| localization_valid | boolean | Indicates whether localization is valid based on centerline geometry completeness, projection computability, and the spatial relationship between the vehicle and lane. |
| lane_segment_progress | number | Normalized longitudinal progress of the vehicle along the lane centerline, with 0 at the segment start and 1 at the segment end. |
| longitudinal_position | number | Cumulative distance in meters from the start of the lane segment to the vehicle reference point along the lane centerline. |
| distance_to_centerline | number | Absolute distance in meters from the vehicle reference point to the lane centerline. |
| nearest_centerline_index | integer | Zero-based index of the discrete lane centerline point nearest to the projection of the vehicle reference point. |
| projected_centerline_point | object | Coordinates in meters of the nearest point on the lane centerline to the vehicle reference point. |
Trust
Compliance Statement
Authorization Type
Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use
Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization
No PII, no real company names, simulated scenarios follow industry standards
Compliance System
Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs
FAQ
Frequently Asked Questions
- What localization information does the data include?
- The JSON input contains vehicle pose, lane geometry, and coordinate references. Outputs include the nearest centerline point index, longitudinal position, lateral offset, heading difference, projected centerline point, lane-segment progress, distance to the centerline, and lane-membership and localization-validity flags.
- What units are used for distance, angle, and progress?
- Longitudinal position, lateral offset, distance to the centerline, and projected point coordinates are expressed in meters. Heading difference is in radians and ranges from negative pi to pi. Lane-segment progress is normalized from 0 to 1. The sign of lateral offset follows the lane travel direction and the right-hand coordinate convention.
- Which training and evaluation tasks does it support?
- It supports structured-output supervised fine-tuning, lane-relative localization, vehicle-lane spatial relationship calculation, map matching, and JSON generation. It can also help evaluate whether localization fields and lane-validity decisions meet task requirements.
- How can I access the data and check its license?
- Refer to the product page for the available download or access method, and review the applicable license, usage restrictions, and delivery specifications before use. Confirm the sample volume and coverage in the product listing or delivery documentation.
Can't find the data you need?
Post a request and let data providers reach out to you.
Get this Dataset
High-quality data, ready for your next AI breakthrough.
Cite this Work
@dataset{Mobiusi2026,
title={Lane-Relative Pose Localization Dataset},
author={MOBIUSI INC},
year={2026},
url={https://www.mobiusi.com/datasets/b805c9a663f3db98b2e73a9d6532c78f?dataset_scene_id=1},
urldate={2026-10-27},
keywords={lane-relative localization data,lane centerline localization,automated driving data,lane-level map data,vehicle lateral offset,heading difference calculation,structured JSON data,localization model training},
version={1.0}
}Using this in research? Please cite us.