[360Labs.ai]
{ 360Labs }Case Study
{ Research and Development }· Live

VAJRA

Multi-sensor on-device counter-UAS system with custom-trained visual (YOLOv8n) and acoustic (CNN) deep learning models. Runs entirely on a smartphone with zero network dependency.

VAJRA
// overview

Complete C-UAS pipeline on a single Android device. Three detection modalities: visual (camera + custom-trained YOLOv8n, 12MB), acoustic (microphone + FFT + custom-trained CNN, 129KB), and RF spectrum analysis. All fused into a unified tactical threat display with countermeasure control.

// specs

Specifications

VisualCustom YOLOv8n, 12MB, >25 FPS
AcousticFFT + CNN classifier, 129KB, <100ms
Database8 drone profiles with 22 parameters each
APK38MB total, zero network dependency
// features

Features

  • 01Custom-trained YOLOv8n (nano) on drone-specific imagery at 320x320
  • 02Real-time 1024-point FFT for propeller frequency detection (50-500 Hz)
  • 03CNN classifier on mel spectrograms for 5-class drone type classification
  • 048 drone profiles (DJI Mavic 3, Bayraktar TB2, Shahed-136, FPV Attack, etc.)
  • 05All ML inference runs locally via TensorFlow Lite
// tech stack

Tech Stack

KotlinTensorFlow LiteYOLOv8CameraXFFTAndroid

Interested
in VAJRA?

Let's discuss how this can work for you.

Get in Touch →
// end of case studyVAJRA
Contact