Ultrawide band-based navigation for multi-agents
Seah, Ryan Meng Yong
Date of Issue2019
School of Physical and Mathematical Sciences
This project is about the use of neural networks in the development of a ultra-wideband (UWB) based localization system for micro-UAV swarm. The software in this project was developed in Linux and programmed in Python. The micro-UAV swarm and UWB localization system hardware were purchased from Bitcraze AB. A total of two experiments were conducted to train the neural network. One of the neural networks used two micro-UAV in the training process while the other used three micro-UAV. The experiment conducted using two micro-UAV drones, minimized the maximum positioning error from 20.2 cm to 18.6 cm. Another experiment, conducted with 3 micro-UAV drones, further reduced it to 15.6 cm. The result showed that neural networks can help in the calibration phase of UWB anchors in minimizing the positioning error in UWB localization. This project is important as it demonstrates the use of neural network in minimizing positioning errors in localization problems in real-life applications is possible.
Final Year Project (FYP)