Design and Implementation of a Custom Flight Controller System

Abstract

The project aims to build a custom flight controller system for a quadcopter. It acts as the flight’s brain performing functions like acquiring sensor data from the Inertial Measurement Unit (IMU) to estimate the state of the vehicle which includes it’s orientation, altitude,velocity to control the actuators (motors) using a PID control loop to maintain a stable flight. The analysis is carried out using a Monte Carlo simulation approach, where multiple independent iterations are performed to ensure statistical reliability of the results.

We are building this system from the ground up with the aim of making it modular and also a cost effective option compared to commercial off- the-shelf flight controllers like Pixhawk and Ardupilot by removing unused ports on universal controllers and only keeping those required for the core functionality i,e. to maintain a stable flight. This project also aims to reduce costs by using cost-effective components like the LSM6DSL(6-axisAccelerometer  and Gyrometer) and MLX90394RLD-AAA-000-

 

SP (Magnetometer) together working as the IMU along with other components whilst maintaining performance.

The system integrates high precision sensors like the IMU, Barometer(BMP390). For better accuracy, we use Madgwick filter, a sensor fusion algorithm which combines the high-speed noisy data from the sensors to estimate the 3D orientation of a device by combining data from sensors like gyroscopes, accelerometers, and magnetometers. We are building this flight controller system on a 4-Layer PCB using an STM32F411RETx Microcontroller Unit(MCU) and incorporating all the sensors, external oscillators, and ports on a 60mmx60mm board for a small but effective design. generation wireless communication systems such as 5G and beyond.