Rare and difficult events
Critical situations may appear a handful of times in thousands of hours of driving or operation.
Synthex for Autonomous Systems
Explore configurable scenarios, synthetic perception datasets and model evaluation workflows designed around difficult physical-world conditions.
Built with these teams in mind
Simulation supports development and testing; it does not establish real-world safety or certification.
Industry challenges
Perception systems meet situations that are rare, dangerous or expensive to capture. Controlled scenarios make them repeatable.
Critical situations may appear a handful of times in thousands of hours of driving or operation.
Rain, fog and spray change appearance and sensor returns in ways clear-weather data does not cover.
Night, glare and low contrast reduce what cameras can resolve.
Objects partly hidden, fallen over or in unexpected places break learned assumptions.
Different mounting, calibration and noise change what a model receives.
Teams often cannot say which conditions their test data actually covers.
Fleets, staging and labelling make each new scenario costly to capture.
Core capabilities
Planned capabilities. They connect to the Simulation, Data and Evaluation engines.
Describe actors, conditions and success criteria as structured, versioned scenarios.
Assemble roads, junctions and yards from configurable layouts and assets.
Capture images and labels across controlled variations of each scenario.
Boxes, masks and sensor metadata derived from known scene ground truth.
Test models on hard, rare scenarios and group failures by cause.
Re-run identical scenario suites as models change and compare like for like.
Interactive explorer
Combine an environment, weather, lighting and detection task to see how a perception scenario could be described. Educational illustration: no simulator is executed.
Illustrative Interface — No Simulation Executed
Illustrative Demonstrationintersection.rain.dusk.pedestrian
Scenario summary: Pedestrian Detection in an intersection scene, rain weather, dusk lighting.
Expected annotation types
Example evaluation concerns
Educational illustration. It does not run CARLA, Unreal Engine, Unity or any physics simulation, and it produces no data. Simulation-based testing does not by itself establish real-world safety.
Simulation workflow
The intended loop for scenario-based perception testing. Failure analysis feeds the next round of scenarios.
Capture is conceptual
STEP 01
State the situation, actors and what the perception system must detect.
STEP 02
Load or assemble the environment layout.
STEP 03
Vary agents, positions, weather and lighting within defined ranges.
STEP 04
Record camera and other sensor outputs with exact ground truth.
STEP 05
Validate labels and package the captured data.
STEP 06
Evaluate per scenario, condition and object type.
STEP 07
Group errors by cause and inspect representative frames.
Scenario Refinement · returns to step 01
Failure categories define new, harder scenario variations for the next cycle.
Sensor data concepts
Representations a scenario-based workflow would work with. None is presented as actively supported: no capture pipeline is implemented yet.
Colour images from configurable virtual cameras: placement, field of view, exposure.
Per-pixel distance from the camera, known exactly in simulation.
Point clouds from modelled laser scanners; realism depends heavily on the sensor model.
Pixel-level class and instance labels aligned to each camera frame.
2D and 3D object boxes with class, occlusion and truncation attributes.
Weather, lighting, actor positions and sensor settings recorded per frame.
Edge-case examples
Illustrative edge cases and what a perception team could evaluate in each. Drawings are schematic, not simulator output.
Edge case 01
What to evaluate
Detection rate and box accuracy as visible area shrinks behind parked vehicles.
Edge case 02
What to evaluate
Detection range and confidence as contrast falls with distance.
Edge case 03
What to evaluate
Missed low obstacles outside headlight cones; false alarms from reflections.
Edge case 04
What to evaluate
Tracking many agents crossing paths; confusion between overlapping objects.
Edge case 05
What to evaluate
Behaviour on classes absent from training, such as fallen loads or debris.
Edge case 06
What to evaluate
Detection of people and forklifts in dust, steam or poor lighting.
Technology
Synthex is evaluating established simulation tools as possible integrations or research tools behind a common scenario layer. No integration is complete, and there is no partnership with or endorsement by their owners.
Backend-independent scenario definitions, variation sweeps, capture settings and hand-off to the Data and Evaluation engines.
Open-source simulator for autonomous driving research with configurable maps, traffic and sensors.
Real-time 3D engine from Epic Games, suited to photorealistic environments.
Real-time 3D platform with tooling usable for synthetic data generation.
Robotics-oriented simulators such as Gazebo may suit indoor and industrial scenarios.
CARLA, Unreal Engine, Unity and Gazebo are names or trademarks of their respective owners and are mentioned only to describe technologies under consideration.
Pilot positioning
Example answers to the questions a perception or simulation lead would ask. Deliverables are examples of what a scoped pilot could produce — not an operational service.
Autonomous systems pilot
We are interested in working with perception and simulation teams on scenario design, synthetic data and edge-case evaluation.
Simulation-based testing can help find weaknesses, but it does not by itself establish real-world safety or meet any certification requirement.