RadiationExplorer

What it is. An Unreal Engine plugin for scientific visualization and analysis of radiation data — load volumetric simulation output, interpolate values at arbitrary points, and render isocontours and particle visualizations in a real-time 3D environment. Applications: radiation-safety analysis, nuclear-engineering simulations, scientific research, and AR/MR walkthroughs of dose fields.

Why. Radiation transport simulations produce dense scalar/vector fields that are hard to interpret without being inside them. Standard desktop tools (ParaView, VisIt) show the field; an Unreal-based viewer lets you walk through it, measure accumulated dose along a path, and explore design alternatives interactively, including in XR.

Lineage: VIPER → RadiationExplorer

RadiationExplorer is the production descendant of VIPER (Virtual Interaction with Physics Enhanced Reality), the earlier AR-based system for visualizing ionizing radiation data. VIPER was presented at the ANS 2022 Annual Meeting (DOI 10.13182/T126-37927) and is the subject of US patent applications 63/438,888 and 18/134,649 (2023), held by UT-Battelle (Oak Ridge National Laboratory). Scott Greenwood is a named inventor on both.

How it ingests data

The data pipeline is a three-step VTK → ParaView → Unreal flow, with a batch workflow that runs the whole thing in one call:

Step	What	How
1. File structure	organize raw VTK files into a folder hierarchy keyed by location	python pipeline_helper_createFileStructure.py
2. Isocontours	generate .obj isocontour meshes from each VTK via ParaView	pvpython.exe pipeline_helper_createIsocontours.py
3. Load to UE	import the generated assets into Unreal	py pipeline_helper_loadData.py (run from the UE editor)
all three	batch workflow	py pipeline_batch_workflow.py

A globalCoordinates.csv maps each simulation origin to the Unreal world origin, so multi-source scenes line up correctly. Two folder conventions are supported: flat (one location) and nested (per-location subfolders). Coarse “pawn” grids — for whole-person dose estimates — can sit alongside fine grids used purely for visual meshes.

Two visualization approaches

Group Visuals — process a structured set of VTKs through the workflow editor function to auto-create base contours, then drop a BP_GroupIsocontourVisualizer referencing the group. Best for static radiation fields you want to show.

Data Visuals — drop a data-source interpolator actor into the scene with the right properties, then add a Blueprint deriving from A_DSI_Accessor to query the interpolated field at any point. Best for visuals or other actors that react to the data — dose-rate readouts, moving sensors, dynamic alarms.

Data Source Manager

A managed-pair system for assembling data + visual pairs in a scene without hand-wiring each one. One BP_DataSourceManager in the level (singleton); on actors that need data exposure, add an AC_ManagedDataSource component, point it at a config row name (e.g. "Corner", "Center"), and on Play the pair spawns automatically and attaches to the owner.

Capability	API
Component-side control	Enable() · Disable() · SetEnabled(bool) · GetPair()
Manager-side bulk control	EnableAll() · DisableAll() · EnableByRowName("Corner") · DisableByRowName("Center") · GetAllPairs() · GetAllEnabledPairs()
Dynamic registration	Register() · Deregister() · RegisterIntegratorActor(actor) · DeregisterIntegratorActor(actor)

Integrator actors sample every registered data source via an AC_PointInterpolator (one per source) and accumulate dose and dose rate in real time. Concretely: a Pawn walking through the scene carries an integrator; the manager wires it to every active source; the Pawn always knows its accumulated exposure.

Three canonical workflows:

1. Static scene — components placed at edit time; pairs spawn on Play.
2. Step-by-step reveal — components start disabled; gameplay calls Enable() to progressively unfold the visualization.
3. Dynamic spawning — spawn an actor at runtime, add the component via Blueprint, call Register() — the pair appears without a pre-existing slot.