Acron

Acron for Industry & Research

Real-Time 3D.
Python-Native.

A production-grade 3D engine that speaks your language. Connect sensor data, ML models, physics simulations, and real-time visualisation — all in Python. No game-industry overhead, no bloated toolchain.

Digital Twins Simulation Robotics Research Environments Data Visualisation
Talk to Enterprise Team Try Free

Why Acron

Your Data Pipeline.
In the Engine.

Industrial simulation platforms lock you into rigid connectors and expensive per-seat licenses. Acron is Python-native — you can integrate any data source, any model, and any protocol directly in Python, with C++ available for real-time performance-critical loops.

  • ▪NumPy/SciPy in-engine — sensor arrays, signal processing, control algorithms in the render loop
  • ▪Bullet physics — rigid body, soft body, collision — not a game gimmick, a physics engine
  • ▪Distributed state — multi-node simulation via DO stack; synchronise state across cluster nodes
  • ▪Headless mode — run simulation loops without a display; pipe output to CSV or a live dashboard
  • ▪Deploy freely — no per-seat limits and no vendor approval to ship your simulation
digital_twin.py 
import numpy as np
from acron import Acron, Entity
from your_sensors import read_lidar_frame

app = Acron(headless=False)
factory_floor = Entity(model='factory.glb')

def update():
    # Read live sensor data each frame
    points = read_lidar_frame()   # numpy array
    heatmap = np.mean(points, axis=0)
    factory_floor.set_heatmap_texture(heatmap)
    # Trigger alert if anomaly detected
    if heatmap.max() > THRESHOLD:
        alert_dashboard(heatmap)

app.run()

Applications

What Teams Build

Digital Twins

Live 3D replica of a physical asset — factory floor, building, vehicle — updated in real time from sensor streams. Visualise, analyse, and intervene without touching the physical system.

Robotics Simulation

Simulate robot kinematics, sensor perception, and environment interaction. Train RL policies in simulation, transfer to hardware. Compatible with Gymnasium-style interfaces.

Training Data Generation

Render photorealistic scenes, segmentation masks, depth maps, and optical flow for CV dataset generation. Programmatic scene variation at scale — no artist required per frame.

Distributed Simulation Clusters

Use the Distributed Objects stack to synchronise simulation state across multiple compute nodes. Each node owns a spatial partition; state flows through the cluster automatically.

Scientific Visualisation

Plot NumPy arrays, point clouds, volumetric data, and time-series in an interactive 3D environment. Combine with Jupyter for notebook-driven 3D exploration.

Autonomous Systems Testing

High-fidelity physics and sensor simulation for autonomous vehicle and drone testing. Deterministic replay, parameterised scenario generation, CI integration.

Enterprise & Research

Used by research institutions, simulation studios, and industrial engineering teams.

Contact Enterprise Try Free First →