This repository includes schema definitions for serialization.
This repository contains a 3D drone detection model designed for target drone detection, tracking, position estimation, and velocity estimation.
The code in this repository provides the following functionalities:
-
Target Drone Detection:
- Utilizes advanced computer vision techniques and deep learning to identify and locate drones in a 3D space.
-
Target Drone Tracking:
- Implements tracking algorithms to follow the movement of identified drones over time.
-
Target Drone Position Estimation:
- Utilizes the detected features to estimate the precise position of the target drone in a 3D coordinate system.
-
Target Drone Velocity Estimation:
- Calculates the velocity of the target drone based on its movement patterns.
Follow the steps below to use the 3D drone detection model:
-
Install MiniConda3
- Download Miniconda package here
cd ~/Downloads bash Miniconda3-latest-Linux-aarch64.sh
-
Install Ecal In a new terminal
sudo add-apt-repository ppa:ecal/ecal-5.11 sudo apt-get update sudo apt-get install ecal sudo apt install python3-ecal5
-
Create Conda Environment
cd ecal-common/python conda create -n vilota python=3.9 -
Install Dependencies:
conda activate vilota pip install matplotlib==3.5.1 numpy==1.25.0 pycapnp==1.3.0 ultralytics==8.0.220 protobuf
-
Running Detection Program:
python3 test_detection3d.py
-
If you face this error
import ecal.core.core as ecal_core ModuleNotFoundError: No module named 'ecal'- Open test_detection3d.py
- Add these lines to the top of the python file
import sys sys.path.append('/usr/local/lib/python3.8/dist-packages') print(sys.path)
This project is licensed under the MIT License. Feel free to use, modify, and distribute the code for both non-commercial and commercial purposes.
Happy coding!
