Scribd
Upload
17 views13 pages

ENG 4530 Lab 2

The document outlines the objectives and tasks for Lab 2 of ENG 4530 at Middle Tennessee State University, focusing on designing a PID controller for a DJI Tello drone and implementing face detection using OpenCV and deep learning. It details the PID tuning process, including specific coefficients for yaw, forward/backward, and up/down movements, and emphasizes the importance of understanding the code for face tracking. The report requires a structured delivery including an introduction, task descriptions, and a conclusion summarizing the lab's achievements.

Uploaded by

Gideon Tyree
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
17 views13 pages

ENG 4530 Lab 2

The document outlines the objectives and tasks for Lab 2 of ENG 4530 at Middle Tennessee State University, focusing on designing a PID controller for a DJI Tello drone and implementing face detection using OpenCV and deep learning. It details the PID tuning process, including specific coefficients for yaw, forward/backward, and up/down movements, and emphasizes the importance of understanding the code for face tracking. The report requires a structured delivery including an introduction, task descriptions, and a conclusion summarizing the lab's achievements.

Uploaded by

Gideon Tyree
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 13

ENG 4530 Lab 2

Middle Tennessee State University


Objectives
• In this lab, you will learn how to design a PID control to control DJI
Tello drone.
• You will learn how to design a face detection computer vision tool
• You will learn how to design and fine tune the PID coefficient to have
drone track face.
Tasks
• DJI Tello drone face tracking.
• Design OpenCV based face detection
• Design Deep learning based face detection
• You need to run main.py attached in the code section
• Note That: Main.py, utilities.py, and haarcascade_frontalface_default should be in the
same folder.
• Design the PID controller using OpenCV based face detection track face, You
mainly only use P and D to control the drone.
Tasks: Main.py
• First coefficient is P, second coefficient is D, Third is I – not
implemented here.
• pid_yaw = [0.5, 0.5, 0]
• pid_fb = [0.1, 0.1, 0]
• pid_ud = [0.556, 0, 0]
Report: Introduction
Please review this video
• What is a PID Controller? | DigiKey – YouTube
Report: Introduction

PID Pseudo Code


Wath this Video: PIDs Simplified - YouTube
Report: Introduction

• What happens if P too big or too low?

Too big P Too small P


Report: Introduction (P +D)
Report: Introduction

Without I With I
Tasks
• PD Tunning process
• Start with low P and low D.
• Increase P, until you get some overshoot and oscillation.
• Then increase D to remove the oscillation.
YAW Given D P Value Response Time

Task A
Value
0.5 0.5 2s
0.6 0.5 Oscillation
0.6 0.6 Oscillation
Given D 0.6 0.7 1s
0.6 0.65 Oscillation

Given D Value P Value Response Time

0.7 0.1 0.62 s

Given P Value D Value Response Time

Given P
Given P Value D Value Response Time
Task B
• Please explain how the Face Tracking works.
• You can achieve by reading the code and understand the code.
• I want you comment on the code and understand each blocks.
• It will be ideal if you are able to change part of the code to test the
changing of face tracking performance.
Delivery
• Introduction
• Describe objectives of the lab
• Describe equipment used in the lab
• Lab Tasks
• Describe results of each task of the lab
• Conclusion
• Revisit each task to conclude them
• Conclude achievements of the lab

You might also like