matlab-driving-data-importer
Import recorded driving sensor data (GPS, camera, lidar, actor tracks, lanes) into scenariobuilder.* objects (GPSData, CameraData, LidarData, ActorTrackData, Trajectory, laneData) and run preprocessing — synchronize, offset correction, crop, normalizeTimestamps, convertTimestamps. Also: compute actor tracks from lidar when no annotations exist, attach camera/lidar mounting + intrinsics, export to MAT/workspace/timetable/script. Use for raw driving dataset files (KITTI, nuScenes, Waymo, Pandaset, ROS/ROS2 bags, .mat, .csv, .mp4) or driving/vehicle/sensor logs that need wrapping. drivingLogAnaly
npx skills add matlab/matlab-agentic-toolkit --skill matlab-driving-data-importer --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
The skill loads raw driving sensor data into scenariobuilder data objects (GPSData, CameraData, LidarData, ActorTrackData, Trajectory, laneData) and provides a CLI for preprocessing steps: synchronize, offset correction, crop, normalizeTimestamps, convertTimestamps. It can compute actor tracks from lidar when no annotations exist and attach camera/lidar mounting, intrinsics, and export to MAT/workspace/timetable/script. It supports raw driving datasets (KITTI, nuScenes, Waymo, Pandaset, ROS/ROS2 bags, .mat, .csv, .mp4) or driving/vehicle/sensor logs. DLA (drivingLogAnalyzer) is opt-in only and must be requested explicitly or triggered by a sensor-data problem; it is not auto-launched after wrapping. After wrapping, the next canonical step is to hand off to matlab-scenario-builder for scenario export.
How it works
- Step 1: Load data into GPSData with timestamps, latitude, longitude, altitude, followed by convertTimestamps(..., 'numeric') and normalizeTimestamps to establish a numeric timeline starting at t=0.
- Step 2: If per-frame 3D annotations exist, map them to ActorTrackData; if annotations are world-frame, transform to ego-relative coordinates.
- Step 3: Load CameraData by matching timestamps with image files and attach metadata like Name.
- Step 4: Load LidarData by matching timestamps to sensor files and assign a Name per sensor for multi-sensor setups.
- Step 5: Synchronize all sensors (convertTimestamps where needed, obtain a time reference, and call synchronize with the relevant data objects).
- Step 6: Launch DLA only when explicitly requested or when a sensor-data problem is reported; otherwise, proceed to matlab-scenario-builder.
- If there are no annotations, the system will compute actor tracks from lidar data.
When to use it
- When you have raw driving dataset files (KITTI, nuScenes, Waymo, Pandaset, ROS/ROS2 bags, .mat, .csv, .xls, video) and need them loaded into scenariobuilder.* objects.
- When you need multi-sensor synchronization, offsets, cropping, and timestamp normalization across GPS, lidar, and camera data.
- When you need to attach mounting/intrinsics information and export to MAT/workspace/timetable/script.
- When there are no pre-computed annotations and actor tracks must be computed from lidar data.
- When you want to inspect or debug with DLA, and you explicitly request it or report a sensor-data problem.
What it can touch
- scenariobuilder.GPSData
- scenariobuilder.CameraData
- scenariobuilder.LidarData
- scenariobuilder.ActorTrackData
- MATLAB workspace (base workspace for wrapped objects)
- File I/O for MAT/workspace/timetable/script exports
Caveats
- DLA is opt-in only; it will not be launched automatically after wrapping. Use explicit requests like 'DLA', 'drivingLogAnalyzer', or 'open in DLA' to invoke.
- If pre-computed annotations exist, actor tracks may be mapped directly; otherwise, tracks may be computed from lidar data.
- The workflow assumes the next step after wrapping is matlab-scenario-builder for scenario generation and export.
# Driving Data Importer This skill loads raw driving sensor data into `scenariobuilder.*` objects (`GPSData`, `CameraData`, `LidarData`, `ActorTrackData`, `Trajectory`, `laneData`) and provides the CLI for every preprocessing step DLA exposes (sync, crop, offset, normalize, convert timestamps). It also covers the `drivingLogAnalyzer` (DLA) app **as an opt-in inspection tool** — see *Rule 0* below; DLA is never a default step. After wrapping is done, the canonical next move is `matlab-scenario-builder` (trajectory smoothing, scene/scenario generation, lane localization, RoadRunner / drivingScenario / OpenSCENARIO / OpenDRIVE / OpenCRG / Unreal export). ## Rule 0 — DLA is opt-in only (HARD RULE, READ FIRST) **Never call `drivingLogAnalyzer` unless the user explicitly asks for it or has reported a sensor-data problem DLA is built to debug.** Auto-launching DLA after wrapping data is a first-attempt-success failure: it stalls the user (a UI app forces context-switch, scrub, click, confirm) and signals that the agent is not confident the import worked. **When DLA IS allowed — only these two cases:** 1. **Explicit user request** — user types `DLA`, `drivingLogAnalyzer`, "open in DLA", "i
- Rule 0 — DLA is opt-in only (HARD RULE, READ FIRST)
- When to Use
- When NOT to Use
- IMPORTANT — Execution Rules
- Rule 1: Inspect Before Importing
- Rule 2: Check for Existing Annotations Before Computing Tracks
- Rule 3: Understand Dataset Structure Types
- Rule 4: Validate Coordinate Frames
- Rule 5: Report Data Summary to User
- Common Dataset Formats
- GPS / GNSS / IMU
- Lidar Point Clouds
- Camera Images
- Calibration
What does the matlab-driving-data-importer skill do?
Import recorded driving sensor data (GPS, camera, lidar, actor tracks, lanes) into scenariobuilder.* objects (GPSData, CameraData, LidarData, ActorTrackData, Trajectory, laneData) and run preprocessing — synchronize, offset correction, crop, normalizeTimestamps, convertTimestamps. Also: compute actor tracks from lidar when no annotations exist, attach camera/lidar mounting + intrinsics, export to MAT/workspace/timetable/script. Use for raw driving dataset files (KITTI, nuScenes, Waymo, Pandaset, ROS/ROS2 bags, .mat, .csv, .mp4) or driving/vehicle/sensor logs that need wrapping. drivingLogAnaly
How do I install it?
Run `npx skills add matlab/matlab-agentic-toolkit --skill matlab-driving-data-importer --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From matlab/matlab-agentic-toolkit, a repository with 868 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.