Radar Development Framework

[final raptor]

Overview

• English: LiDAR module of the Radar Development Framework — a stable scanner core, visualization layer and demo controls for quick integration and debugging of point-cloud features in Arma Reforger.

Key Features

• Pluggable sampling strategies (RDF_*SampleStrategy: uniform, hemisphere, conical, stratified, scanline).
• Replaceable color/size strategies (RDF_LidarColorStrategy, RDF_IndexColorStrategy).
• Visualization: points and segmented rays via RDF_LidarVisualizer; obtain samples with GetLastSamples() for export.
• Safe settings validation (RDF_LidarSettings.Validate()) and sample self-checks (lidar_sample_checks.c).

Quick Start

• Enable demo: RDF_LidarAutoRunner.SetDemoEnabled(true)
• Set sampling: scanner.SetSampleStrategy(new RDF_ConicalSampleStrategy(25.0));
• Export: ref array<ref RDF_LidarSample> s = visual.GetLastSamples();

Notes & Tips

English: Keep m_RayCount/m_RaySegments reasonable for performance; see README.md and DEVELOPMENT.md for examples and extension guidance.

License & Contact

License: Apache-2.0. Contact: [e-mail].
Github:https://github.com/ViVi141/Radar-Development-Framework

[keen wren]

EWS potentially? would be cool to spot helis from far away

[final raptor]

LiDAR can indeed detect any colliding object (except bullets and shells?), but using lidar for detection is too expensive and unsuitable; I plan to build an electromagnetic wave radar system that can detect phase, attenuation, specular reflection, and diffuse reflection.

[final raptor]

I don't think anyone could tell that it was actually a tree.

However, traditional AI can definitely do it, because it can use this framework to perceive spatial changes in real time, including the distance and coordinates of each point, object names, and materials.

[zenith swallow]

Does this amount of ray cast not kill FPS, this is very impressive though

[final raptor]

Not visualizing these rays won't affect the frame rate, but it's useful for debugging.If 4096 rays are drawn every 0.2 seconds, the frame rate will drop to 7 FPS.

[glacial oyster]

nice work but i was hoping for some radar for antiair to track flying things in the sky 😄 and display at some UI or screen 😄

[autumn walrus]

this mod is Lidar for automotive

[final raptor]

Lidar is indeed more suitable for low-speed moving objects on the ground, such as characters and vehicles. Due to its characteristics, lidar can only detect in a straight line and is not good at detecting flying objects that are obstructed.

[final raptor]

Of course, it is also suitable for AI pathfinding.

[final raptor]

Updated to version 0.3.0. Enhanced network synchronization capabilities of the LiDAR module, adding new network APIs and components to support reliable transmission in server-authoritative architectures and high-ray-count scenarios. Refactored API, introducing asynchronous scanning functionality. Optimized demo configuration, supporting point cloud-only mode; improved CSV import/export, enhancing serialization and compression efficiency. Updated documentation, providing change notes and networked LiDAR sample code, and optimized scanner and visualization logic.

[final raptor]

I'm currently working on a LiDAR vehicle prototype and have added some fun tools.I'm currently working on a LiDAR vehicle prototype and have added some fun tools: it can export CSV files to the local disk and play them.

[final raptor]

Example of LiDAR vehicles: https://reforger.armaplatform.com/workshop/6896E11A7CBD40B1-ExampleofLiDARvehicles andThe LiDAR module has been upgraded to version 0.4.2. It has been fixed and optimized, and local scope variables are now used to reduce temporary object allocation. The network serialization logic has been optimized to improve the efficiency of fragment reception and assembly. The asynchronous scheduling strategy has been adjusted, and the rendering array has been pre-allocated to reduce memory jitter.

[carmine vessel]

Looks pretty cool. 😎

I just don't fully get it...

[scenic osprey]

shooting all thoes rays must be costly

[autumn walrus]

it is a simulation for real-life Lidar sensor

[carmine vessel]

Yeah i have to do some research.

[final raptor]

1024 raycasts don't incur significant performance overhead; the performance bottleneck lies in rendering the rays and the CSV output. However, I believe blindly increasing the number of rays is a mistake. Developers must customize scanning strategies for their work, such as adaptive, LOD, and cone-shaped methods.

This demo didn't run well on my 4060 laptop. I miss my 7900XT from day one.

[final raptor]

Under development: Visual rendering has been optimized, increasing the FPS of the vehicle-mounted LiDAR demo from 6 to 65 (on a 4060 laptop), shifting some overhead from the CPU to the GPU.

[west trellis]

Could be interesting to integrate this into the ZSU Shilka

[final raptor]

Yes. This is a good solution for unguided bullets.

[west trellis]

The zsu has a radar prop on it. Having it actually do something might be cool. Rather than it being pointless.

[final raptor]

Yes, although only lidar has been implemented at present, targeting within the field of vision is already available. The real radar is still under development.

[final raptor]

You can ignore the previous message. This is a development progress announcement, but it's too long. I'll send a more appropriate one when I have time.

[final raptor]

We are pleased to announce that the initial preparations for the electromagnetic wave radar system have been completed. The required engine APIs for the electromagnetic wave radar system have been validated, achieving 100% completion. Furthermore, an implementation plan for the electromagnetic wave radar system has been developed. This plan aims to extend the existing LiDAR ray tracing framework into a complete electromagnetic wave radar simulation system, incorporating realistic radar physics characteristics, including electromagnetic wave propagation, signal attenuation, Doppler effect, and target reflection characteristics. The system will maintain compatibility with existing LiDAR architectures and provide independent radar modules for users to choose from.

[final raptor]

In testing, this was not impossible.

[carmine vessel]

So the radar is working with data from the LiDAR?
A'm i getting this right?

[final raptor]

No—radar does not use lidar data; they are implemented separately in this project. The only similarity is that they both use ray detection, but the calculation and interpretation methods are completely different, and radar ray detection can penetrate objects.

[carmine vessel]

Alright got it.
Thank you. 😉

[hollow cosmos]

So will you make AA missles with this framework?

Also how does perfomance look when radar is on?

[final raptor]

It's uncertain; there are currently no plans to develop anti-aircraft missiles, and our current focus is on this project.

[uncut heart]

Looks sick as fuck man great work

[final raptor]

This thing is extremely difficult to make. It's almost like murdering my brain cells. Although the demo has been completed, there are still many areas that need optimization.

[final raptor]

Using rays to simulate the effect of electromagnetic waves doesn't meet my expectations. I need to study more or force a wave to come out?

[uncut heart]

I bet and I feel that hopefully u can figure it out

[scenic osprey]

you probability and statistics, dont do fully ray casts, cast less raycast and try to figure our chance of detection

[keen wren]

is it possible to make guns target stuff based off of radar data? Like a CIWS or a Gepard

[final raptor]

Yes, it can lock onto and track a specific object among multiple scanned objects.However, it is still under development.

[final raptor]

📡 Radar Development Framework — Update Log

Update Date: 2026-02-20

---

🆕 New Features

Electromagnetic Radar System (First Full Release)

• Complete physical simulation pipeline: FSPL propagation loss, atmospheric attenuation, rain attenuation, radar equation (SNR), RCS target cross-section model

• Doppler frequency shift processing + MTI moving target indication (filtering stationary clutter)

• CA-CFAR / OS-CFAR constant false alarm rate detection (built-in offline lookup table, automatic loading)

• 4 operating modes: Pulse / Continuous Wave (CW) / Frequency Modulated Continuous Wave (FMCW) / Phased Array

• On-chip PPI scan circle HUD (pure script CanvasWidget, green theme)

• Target classification system: Entity pre-classification (structural hierarchy / name heuristic) + target classifier

• SAR Synthetic Aperture Radar Processor (Aperture Accumulation)

• ECM Electronic Countermeasures Jamming Model

• PPI Display / A-Scope Range Amplitude Map / World Pillar Markers / ASCII Console Map

• Complete Demo System (AutoRunner / Demo Cycler / StatsHandler)

EM Voxel Field System (Phase 1–5 Complete)

• 3D Electromagnetic Field Simulation Voxel Mesh, Supports Signal Descriptor and Pulse Feature Recording/Estimation

• Passive Sensor (Listening to Signals in the Field)

• Multi-user Network Synchronization (Server Authority)

• Debugging Visualization

LiDAR HUD (New)

• On-chip PPI Scan HUD (Blue Theme), can be used independently or connected to AutoRunner for automatic refresh

---

✨ Improvements

LiDAR

• New public static APIs for HUD: IsVisible() / FeedSamples() / AttachToAutoRunner() / DetachFromAutoRunner() supports direct HUD driving outside the Demo system.
• Added a networked LiDAR configuration example (RDF_LidarNetworkSetupExample)
• Scan sampling extension: Ray hits now carry entity references and surface material information.
• Improved Sweep rotation strategy: Highly uniform distribution, eliminating clustering.
• AutoRunner memory optimization: Proactively releases shape and sample references to prevent leakage during long-term operation.

Radar HUD

• Added 4 public static APIs, with interfaces symmetrical to the LiDAR HUD.
• Updated target color encoding for helicopter radar mode, making classification display more intuitive.

---

🐛 Bug Fixes

• Fixed an error in the z-value calculation of the upper hemisphere sampling strategy (RDF_HemisphereSampleStrategy), causing some rays to deviate from the upper hemisphere.
• Fixed degenerate results in the scanline sampling strategy (RDF_ScanlineSampleStrategy) under extreme parameters.
• Fixed an issue where m_DisplayRange was unexpectedly overridden by the AutoRunner range when the LiDAR HUD was called independently.

---

📚 Documentation

Added RADAR_API.md (complete radar API reference), RADAR_TUTORIAL.md (14-section usage tutorial), and CHANGELOG.md; cleaned up all internal planning documents, retaining only reference materials for developers.

---

⚠️ The radar module and demo are still under development and should not be used in production environments!!! They will remain in the early development stage for a long time and are currently completely unstable and unusable. The LiDAR module is relatively stable.

Repository: https://github.com/ViVi141/Radar-Development-Framework | Contact: [e-mail]

[final raptor]

Under development: The LiDAR boasts impressive scanning accuracy, correctly identifying vehicles, people, and buildings, even down to individual vehicle components and personnel equipment. It filters out irrelevant information such as terrain, making it well-suited for use with armed vehicles.

[final raptor]

Under development: Intelligent object filtering is applied to a vehicle example to retain valid objects (volume greater than 0.0001 cubic meters and not the ground itself). It can accurately identify people, buildings, and landmines, and allows users to zoom the HUD scale using a button. When people crouch and get too close to landmines, the lidar will fail to detect them (this is entirely based on light detection, not hard logic).

[final raptor]

Change log version 0.6.1
Implement Target Filtering Modes for LiDAR and Radar Systems

• Introduced the ETraceTargetMode enumeration to define target filtering options: Terrain Only, All (Terrain + Entities), and Entities Only.

• Updated RDF_LidarSettings and RDF_RadarSettings to use the new target filtering modes, which affects ray casting behavior.

• Enhanced RDF_LidarDemoConfig to include target filtering mode configuration.

• Modified the demo script to support setting and validating target filtering modes during initialization.

• Updated documentation to reflect the changes to target filtering mode functionality and its impact on scan behavior.

[final raptor]

In the latest LiDAR vehicle example, with settings of 4096 ray detections, a refresh rate of 10 times per second, and ignoring terrain while only recognizing valid entities, the frame rate difference between the previous scan (i.e., not yet on the vehicle, not yet scanning) and the previous scan (i.e., during scanning) is 5 FPS. This frame rate fluctuation is likely due to differences in CPU performance and network volatility.

The above test scenarios are run on the server and the calculations are performed by the server.

[crimson tusk]

is there a way to implement the radar in an MFD ? in a helicopter for ex

[final raptor]

The code provides a demo with a HUD, but it's not drawn on the pip; it's just a simple overlay. Generally speaking, it can be achieved, but I don't know how to write it. This needs to be handed over to the developers using this framework.

The current HUD can be found in the previous chat records.

[crimson tusk]

Alr thank you

[final raptor]

https://youtu.be/6v9JsMrDF00?si=6fZXqBmiz7sdLP_7

[final raptor]

Smoke hinders laser detection.

Currently preparing for the Laser anti-sniper system because I've heard many players are exhausted from being harassed by snipers; it's time to punish them.

[sour pumice]

This is very cool

[crimson tusk]

https://cdn.discordapp.com/attachments/1370780256763379812/1476337069067604069/6FD0cpf.gif

Perfect

[keen wren]

what about counter artillery radar?

[final raptor]

It is currently unclear whether it is possible to scan artillery shells. If it is possible, the position of the artillery can be calculated based on the trajectory of the shells (it would be too boring to directly obtain the world coordinates of the artillery through the game engine; some fun needs to be added).

[next folio]

Hey viví something I want to bring to your attention, while i understand the limitations going up hills, i believe there is a strong debuff while going down hills which should be a significantly easier, faster process instead as it stands it makes the character very very slow

Also if your character has stamina and hasn’t fatigued himself much perhaps going up hills even steep ones could be a bit of a faster process as well until they begin to fatigue

[final raptor]

1. Currently, going downhill does have a slight speed-increasing effect, but this only applies to gentle slopes. On steeper slopes, it slows you down, forcing you to take small steps to avoid sliding down.

2. The speed limit on uphill slopes is indeed a bit strict at the moment. However, a five-second buffer period is provided between flat ground and an uphill section to prevent you from being glued to the ground as soon as you reach the top. I'm currently looking for a better solution and model.

Thank you for your feedback, but please use the correct channel next time.

Have you ever tried a system where AI uses the same stamina as a player? What did you think of it?

[next folio]

I think it’s good nice to see I think though the first earlier iteration of this mod was the most realistic since it gave players more leeway to be mobile for much longer as well I think now it has become a stricter extension of armas stamina system

Humans are endurance creatures and we’re meant to be playing as physically fit soldiers so seeing that our stamina is so limited and treats us like we’re carrying 300kg is a bit unfortunate but I hope for the best in the end

[final raptor]

So you think the current areas for improvement are the speed impact caused by the slope and the other stamina consumption? I'm a mod developer, so it's easy for me to test according to the optimal logic of the mod and get good test results, but I don't really understand the thoughts of ordinary players.

[next folio]

Well, I tested the mod with various different loads on my person, whether it was 40kg or 35kg or 10kg your speed might’ve taken a hit and even some endurance but nonetheless you could jog across the airfield as if it were irl training, I think you should adjust the mod to host a short burst full speed tactical sprint by double clicking the shift button in case of a dangerous situation and just a regular jog to maintain speed and the heavier you are the lower the stamina, going up stairs shouldn’t be affected by weight either unless insanely heavy, hills are fine but the speed should be quicker and going down hills since you’re almost at the will of gravity moving down a steep hill should be much faster and easier

[final raptor]

Yes, I also think that double-clicking shift should reflect very fast movement speed and very high stamina consumption. I don't know if players will accept this change, because most players think that double-clicking shift is for long-distance marching. Holding down shift in all situations leads to excessive stamina consumption, which makes players dissatisfied and makes it difficult to determine whether players simply need faster movement speed for marching or for chasing and escaping.

[next folio]

I think currently the players I play with will be very happy to see this change

[final raptor]

It has been quite some time since the last update regarding this mod. We are currently restructuring the design of the electromagnetic wave radar system and are testing the reflection characteristics of various surface materials under both electromagnetic wave and LiDAR illumination.