A systematic approach to the Design, Implementation and Visualisation of a Quadcopter Controller for Autonomous Navigation

This research aims to develop a robust control system for quadcopters by addressing key objectives: creating an analytical mathematical model, extracting physical parameter values, developing a rapid 3D visualization platform on MATLAB, and implementing a robust controller with data and sensor fusion. Quadcopters are a type of agile, multi-rotor unmanned aerial system (UAS) that utilize four rotors for lift, offering a cost-effective and reliable alternative to traditional helicopters.
The research successfully develops an analytical mathematical model, initially analyzed through graphical representation on MATLAB/Simulink software. The non-linear equations were linearized and represented in a state-space model, simplifying the model and enabling the deployment of simpler control and filtering models. To test the control system's robustness, Gaussian white noise was added to the system and filtered using various techniques. Extracting physical parameters using off-the-shelf instruments can be expensive, so alternative techniques and equipment were investigated to obtain the best outcome. After acquiring the parameters, they were integrated into the mathematical model.
To simplify system validation, the research highlights the benefits of a rapid prototyping 3D model for analysis. Various approaches were adopted to identify anomalies within the system and build a robust control system before deployment onto hardware. Initially, an Atmega328P microcontroller was used for in-flight testing, communicating between ground stations (MATLAB/Simulink) in real-time via Bluetooth. Later, a more powerful Raspberry Pi 3 B+ was utilized as an onboard computer. A robust wireless communication system, considering noise filtering and error-checking methods, was successfully implemented using both Atmega328P and RPi 3B+. This included Bluetooth wireless communication and later, WiFi for wireless communication and integration of sensors such as a camera for visual feedback.
In conclusion, this research primarily focuses on quadcopter simulation and performance evaluation, providing a solid foundation for obtaining physical properties, control models, and sensor filtering. The developed control system demonstrates potential for improving quadcopter reliability and performance, with the analytical mathematical model, rapid 3D visualization, and robust wireless communication system aiding in the development and validation of quadcopter control systems.