DRONE DESIGN AND IMPLEMENTATION

Executive Project Report 

 

Mentors

 

Abhinith D

Prateek 

Jobin

 

Mentees 

 

G Ram Rithesh Reddy

Akanksh Vishwan Patil

 

Aim and Introduction
Drone technology is rapidly advancing, becoming more efficient, versatile, and affordable. Innovations in battery life, autonomous navigation, and camera quality enhance their utility in various fields. Drones offer unique aerial perspectives in photography and filmmaking, revolutionizing cinematography and media production. They also inspect infrastructure like bridges, power lines, and pipelines, reducing downtime and enhancing safety. This project aims to create a Drone system and flight controller from scratch, with applications in various surveillance and security fields.

Technologies Used

Matlab Simulink

Ansys Fluent 

STM32 Microcontroller

Mpu6050

Ultrasonic Sensor

Literature Survey
MATLAB Simulink
MATLAB Simulink is a graphical programming environment for modeling, simulating, and analyzing dynamic systems. In drone development, it helps simulate the drone’s flight behavior, tune the PID (Proportional-Integral-Derivative) controller for stability, and test different control strategies. Simulink’s embedded coder can also be used to automatically generate C code from the model, which can be directly deployed on microcontrollers, streamlining the software development process for real-time control.

ANSYS Fluent
ANSYS Fluent is a powerful Computational Fluid Dynamics (CFD) tool that simulates fluid flow, heat transfer, and aerodynamics. For drones, it helps analyze airflow through and around the propellers to estimate the generated thrust. This is used to decide on the rpm of the propellor, improving the drone’s lift, efficiency, and overall performance. Fluent provides detailed visualizations and data that guide aerodynamic and structural design decisions.

 

Methodology

Computational Fluid Dynamics

We ran a CFD simulation on a single rotor under hover conditions to determine the thrust and propellor rpm required for a stable hover in our drone. The goal was to understand how the air flows through the propeller and verify if it could generate enough thrust.

We modeled the rotor with a 15-inch diameter using a twisted blade design — the pitch starts at 20° at the root and gradually drops to 2° at the tip
We decided on 6000 RPM and assumed a very low forward inflow velocity of 0.063 m/s, simulating a near-perfect hover. This gave us an advance ratio (J) of 0.001.

Using these parameters, we set up a CFD simulation where we let air flow over the propeller inside a cylindrical domain, mimicking hover conditions. The solver captured details like pressure differences, flow separation, and wake patterns.

We measured a steady thrust of around 18.99 N from the simulation, which matched our expectations for maintaining a stable hover.

 

PID Controller

We started by building a linear model of the drone’s dynamics — a simplified version that helped us understand how the drone reacts to small changes. This gave us a good starting point for designing our control system.

Then, we moved on to a non-linear model to better capture how the drone behaves in real life, with environmental factors such as wind or during more aggressive movements.

Using this model, we set up the control system and feedback loop. At the heart of this is the PID controller, which constantly adjusts the drone's inputs (like motor speeds) based on the difference between where it is and where we want it to be.

The system relies on live data from our sensors — mainly the IMU, GPS, and barometer. These give the controller the feedback it needs to make fast, accurate adjustments during flight.

This loop of sensing, adjusting, and flying keeps the drone stable and responsive, even when the environment gets unpredictable.

 

Implementation
 

Simulink Virtual Environment Simulation 

 

Pid controller response

 

MPU6050 Sensor
 

The MPU6050 is a widely used MEMS (Micro-Electro-Mechanical Systems) sensor that integrates a 3-axis gyroscope and a 3-axis accelerometer on a single chip. It provides precise measurements of angular velocity and linear acceleration, making it ideal for motion tracking and orientation estimation in embedded systems. The sensor communicates via the I²C protocol, enabling efficient data transfer with microcontrollers like the STM32F11. In drone applications, the MPU6050 is essential for capturing real-time orientation data, which is critical for implementing stable flight control algorithms such as PID.

 

HCSR04 UltraSonic Sensor

 

The HC-SR04 is an ultrasonic distance sensor commonly used for non-contact distance measurement. It operates by emitting an ultrasonic pulse through its transmitter, which reflects off an object and is detected by the receiver. The time taken for the echo to return is used to calculate the distance to the object with high accuracy. The sensor is capable of measuring distances typically in the range of 2 cm to 400 cm, with a resolution of about 3 mm. In the context of drone systems, the HC-SR04 is particularly useful for altitude estimation at low heights, obstacle detection, and supporting hover stabilization in indoor or GPS-denied environments.

 

STM32 Microcontroller

The STM32 microcontroller series, developed by STMicroelectronics, is a family of high-performance, 32-bit microcontrollers based on the ARM Cortex-M core architecture. In particular, the STM32F11 used in this project offers a balance between computational power, low power consumption, and rich peripheral integration. It supports interfaces such as I²C, SPI, UART, and PWM, making it highly suitable for real-time control tasks and sensor interfacing in embedded systems.

In drone applications, the STM32 microcontroller serves as the central control unit, responsible for processing sensor data, executing control algorithms (such as PID), and generating motor control signals. Its deterministic execution, hardware timers, and interrupt-driven architecture make it ideal for time-sensitive operations required in stable and responsive flight control systems.

x `

 

Results

Successfully designed and simulated a PID-based control system for drone stability using MATLAB. Developed and validated a linear and non-linear model of the drone's dynamics, forming the basis for real-time control.

Successful testing of STM32 microcontroller

Conducted CFD simulations of the propeller under hover conditions, confirming that the rotor configuration could generate 18.99 N of thrust, sufficient for stable hover.

The simulations and theoretical work indicate high feasibility and performance reliability, paving the way for physical implementation and further development.
 

Future Scope
 

 

Future Scope
 

Hardware Implementation and real-life gain tuning:

The MATLAB simulations gave us the confidence that our control logic and thrust modeling are sound.However, the actual hardware assembly and flight testing could not be completed due to time constraints.

This will be taken up as a key part of the future scope, focusing on building, integrating, and validating the drone system in real life with real life validation of the controller and tuning of gains.

Camera Integration for Surveillance & Monitoring
 

Mounting RGB or thermal cameras for real-time visuals.

Useful for forest fire monitoring, perimeter security, and live tracking in remote areas.
 

Computer Vision for Autonomous Decision-Making
 

Integration of models for face detection, fire detection, and object recognition.

Enables automatic threat identification and pre-emptive responses (e.g., sending alerts, triggering safety protocols).
 

Swarming Capabilities
 

Deploying multiple drones with coordination for scalable missions.

It is ideal for search and rescue, area surveillance, or large-scale environmental monitoring.

Enhances reliability and adaptability through decentralized control.

Intelligent Autonomous Navigation
 

Combining vision systems with onboard navigation for GPS-denied environments.

Enables obstacle avoidance, path following, and safe landing without human control.

References

Conclusion

 The project brought together a wide array of engineering concepts, including real-time embedded system design using the STM32F11 microcontroller, inertial measurement through the MPU6050, and altitude sensing via the HC-SR04 ultrasonic sensor. The implementation of a PID controller enabled us to regulate the drone’s yaw, pitch, and roll, ensuring stable and responsive flight dynamics. Additionally, thrust estimation through CFD simulations in MATLAB provided critical insights into aerodynamic performance, reinforcing the engineering rigor of the design process.

Through this project, we developed a comprehensive understanding of drone control systems, sensor integration, and feedback loop tuning. We were introduced to practical applications of control theory, sensor fusion, and embedded software development.

Designing the flight controller and drone system from the ground up offered a hands-on experience that was both challenging and rewarding. The iterative process of debugging hardware, tuning control parameters, and validating simulations equipped us with invaluable problem-solving skills and a systems-level perspective on UAV development.

 

Acknowledgements

We would like to extend my heartfelt gratitude to IEEE NITK for their support and sponsorship of our project, without which, this endeavour would not have been possible.

We would also like to thank our mentors for constant support and guiding our hand throughout the project

REFERENCES

Mitch Davis playlist for STM32 based microcontrollers

STM32 standard datasheets for peripheral reference