matlab-scenario-builder

What it does

Generates driving scenes, scenarios, road surfaces, and 3D content from scenariobuilder.* sensor data using Scenario Builder for Automated Driving Toolbox. BUILD, EXPORT, or AUGMENT a virtual scenario/scene/map: ego or actor trajectories, trajectory smoothing, OpenCRG road-surface extraction, 3D asset generation, static-object placement, point-cloud georeferencing + elevation, lane-based ego localization, sensor-fusion tracking, scenario-event extraction (cut-ins, hard brakes, near-misses, ADAS disengagements), or export to RoadRunner, drivingScenario, OpenDRIVE, OpenCRG, OpenSCENARIO, or Unreal Engine. Also: log-to-scenario, scenario harvesting, accident/near-miss reconstruction, SOTIF (ISO 21448) and ISO 26262 scenario coverage, USGS-aerial-lidar augmentation, traffic-sign placement, vision-based vehicle classification for actor assets. NOT for raw-data import or multi-sensor sync/crop/offset/timestamp normalization — route those to matlab-driving-data-importer.

Scenario Builder is a support package that ships with MATLAB as part of the Automated Driving Toolbox and can export outputs to multiple simulator formats for validation and safety analysis. It emphasizes converting recorded driving logs into reproducible simulation scenarios and supporting standards-focused scenario coverage.

Prerequisite: The MATLAB session must have the Scenario Builder for Automated Driving Toolbox support package installed. Verify with:

isfile(which("scenariobuilder.Trajectory"))

How it works

• The skill provides 18 workflows; the core path covers importing raw sensor data, building ego trajectories, and exporting to external targets. It performs data preparation steps, builds trajectories, localizes ego position, adds static objects, and exports to simulators like RoadRunner or drivingScenario objects. It also supports event extraction, sensor fusion tracking, and elevation/georeferencing of point clouds.
• It instructs to validate data, establish a scene with roads (OSM), build ego trajectories, export ego/actor data to RoadRunner, and run a validation/video workflow. It emphasizes the use of explicit checks and conditional branching (e.g., when to export, how to visualize, and which tools to use for verification).
• The workflow table includes specific triggers like requests to export to RoadRunner, add static objects, extract events, or augment with aerial lidar data. It also notes steps for visualization and validation, including the required pre-checks for camera intrinsics and height, and the use of a progress table to track step completion.

When to use it

• When you have recorded driving data (GPS/GNSS, camera, lidar, actor tracks) and need to convert it into a simulation-ready scenario.
• When exporting trajectories or scenarios to RoadRunner, drivingScenario, ASAM OpenSCENARIO/OpenDRIVE/OpenCRG/OpenSCENARIO, or Unreal Engine. Default target is RoadRunner unless explicitly asked to use Driving Scenario Designer or build a drivingScenario object.
• When you need to extract road-surface data (OpenCRG), add elevation, georeference point clouds, localize ego on a map with lane detections, add static objects to a map, augment scenes with aerial lidar, or generate 3D mesh assets from images.
• When you need to extract critical scenario events (cut-ins, hard brakes, near-misses) or perform sensor-fusion tracking for non-ego actors.

What it can touch

• The skill interfaces with multiple tools and formats: RoadRunner, drivingScenario, OpenDRIVE, OpenCRG, OpenSCENARIO, Unreal Engine. It references sensor data inputs like GPS, camera, lidar, and actor tracks, and supports pre-processing steps such as synchronization and offset correction. Specific tool names and file targets are used in the workflow details (e.g., RoadRunner export, drivingScenario object, OpenDRIVE, OpenCRG, and Unreal Engine).

Caveats

• Not for raw-data import or multi-sensor sync/crop/offset/timestamp normalization; those tasks are directed to matlab-driving-data-importer.
• Requires Scenario Builder for Automated Driving Toolbox to be installed.
• License is NOASSERTION.
• Some workflows referenced as on-demand reside in references/ (not expanded here).