https://github.com/matteogianferrari/multilat_sensor_net
Multilateration and Tracking in 3D Space Using a Distributed Network of Simulated Distance Sensors
https://github.com/matteogianferrari/multilat_sensor_net
distributed-algorithm grpc tracking zeromq
Last synced: 7 months ago
JSON representation
Multilateration and Tracking in 3D Space Using a Distributed Network of Simulated Distance Sensors
- Host: GitHub
- URL: https://github.com/matteogianferrari/multilat_sensor_net
- Owner: matteogianferrari
- License: mit
- Created: 2024-12-25T15:36:59.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2025-01-17T16:14:49.000Z (9 months ago)
- Last Synced: 2025-01-17T17:26:48.967Z (9 months ago)
- Topics: distributed-algorithm, grpc, tracking, zeromq
- Language: Python
- Homepage:
- Size: 1.41 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# MultilatSensorNet
> **Multilateration and Tracking in 3D Space Using a Distributed Network of Simulated Distance Sensors**
**MultilatSensorNet** aims to build a **distributed, scalable network of simulated distance sensors** to estimate and track an object's position in 3D space via multilateration. The system comprises four main components—**Node**, **Network**, **Target**, and **Client**—that work together to measure distances, compute multilateration with least squares, and provide real-time object tracking.
---
## Table of Contents
- [Project Overview](#project-overview)
- [Features](#features)
- [System Components](#system-components)
- [Installation](#installation)
- [Usage](#usage)
- [Performance Analysis](#performance-analysis)
- [License](#license)
---
## Project Overview
**MultilatSensorNet** provides an environment in which:
1. A **Target** moves in 3D space following a defined trajectory.
2. Multiple **Node** + **Sensor** pairs measure the Euclidean distance to the Target.
3. A centralized **Network** component aggregates these distances to compute the Target’s position using multilateration with least squares.
4. A **Client** application requests the global Target position from the Network, using a **Tracker** and **KalmanFilter** to track the movement over time.
By utilizing a distributed node-sensor approach, the system aims to be **scalable**, **robust**, and **adaptable** to various use cases.
---
## Features
- **Distributed Sensing**
Multiple Nodes measuring distances in 3D space, communicating results to a central Network.
- **Scalable Multilateration**
Use of a least squares algorithm to estimate the position of the Target from multiple distance measurements.
- **Kalman Filter Integration**
The Client tracks the Target in real time, smoothing measurements via a Kalman Filter.
- **Modular Architecture**
Each component—Node, Network, Target, and Client—can be extended or replaced, enabling flexible experimentation.
---
## System Components
1. **Node (Node + Sensor Packages)**
- Simulated distance sensor that measures the Euclidean distance to the Target in 3D space.
- Handles network requests and communications with the **Network**.
2. **Network (Network + Estimator Packages)**
- Central component managing Nodes registration, distance retrieval, and communication requests.
- Performs multilateration using least squares to compute the Target’s position.
- Communicates computed positions to the **Client**.
3. **Target**
- An object that moves in 3D space according to a defined trajectory.
- Provides distance information to each Node upon request.
4. **Client**
- Requests the global Target position from the Network.
- Maintains a **Tracker** with a **Kalman Filter** for smooth real-time tracking.
---
## Installation
1. **Clone the Repository**
```bash
git clone https://github.com/matteogianferrari/multilat_sensor_net.git
cd multilat_sensor_net
2. **Install Dependencies**
```bash
pip install -r requiremets.txt
## Usage
1. **Start the Network**
```bash
python3 network_main.py
2. **Start the Target Simulation**
```bash
python3 target_main.py
3. **Initialize Nodes (Minimum 3 Nodes)**
```bash
python3 node_main.py --node_id 1 --pos 0 0 0
python3 node_main.py --node_id 2 --pos 4.5 4.5 2.75
python3 node_main.py --node_id 3 --pos 2.8 4.5 0.65
4. **Run the Client**
```bash
python3 client_main.py --verbose
## Performance Analysis
A performance analysis notebook is included in the project. It uses the output file generated by the tracker and compares the computed trajectory with the original target trajectory (input). This helps assess the accuracy and effectiveness of the multilateration estimation and tracking system.
The notebook generates visualizations and error metrics to evaluate the system's performance.
## License
This project is licensed under the MIT License. See the [LICENSE](LICENSE) file for more details.