Advisory Committee Chair
Abidin Yildirim
Advisory Committee Members
Leon Jololian
Mohammad Haider
Document Type
Thesis
Date of Award
2017
Degree Name by School
Master of Science in Electrical Engineering (MSEE) School of Engineering
Abstract
Every year’s thousands of people died in road accidents worldwide. The autonomous vehicles provide a better solution for this kind of situations. The autonomous vehicles move freely in an outer environment and perform different types of tasks according to the condition. The moving object detection in real time and perform the action are always considering important part in autonomous vehicles. To demonstrate the concept of the autonomous vehicle, we design an autonomous all-terrain vehicle model. In this thesis, we used the two types of distance measurement sensors i.e. Ultrasonic and Sharp Infrared sensors. These sensors are connected to the Arduino MEGA microcontroller and programmed with Arduino IDE interface. The Pololu four-wheel drive terrain vehicle is used for the vehicle's platform. We used the motor driver to control four motors. We also used the Raspberry Pi for using open source computer vision application and interface the camera. It is single board credit size computer, which is sufficient for basic computational tasks. We have applied the computer vision algorithms for the real-time object detection in OpenCV software. In this work, we used Logitech web camera to capture and send the images frames to OpenCV software. Once we got the frames, we applied different types of method to detect the moving object from the frames. We used the grayscale conversion, apply a Gaussian blur to blur the frames, delta frames to find the difference between frames, threshold method to find the high intensity and low intensity in frames, find contours to find the contours from the threshold frames and finally draw contours to draw the counter around the detected frame. We saved the values of detected frames as one and zero. One means object detects detected, zero means no object found. Later on, we used detected values to communicate with Arduino MEGA to perform some specific task on the autonomous model. Keywords: Autonomous vehicles, All-terrain vehicle, Arduino MEGA, Raspberry Pi, Computer vision, OpenCV.
Recommended Citation
Singh, Avinash Kumar, "Object Detection And Path Detection System For Autonomous Terrain Vehicles" (2017). All ETDs from UAB. 2981.
https://digitalcommons.library.uab.edu/etd-collection/2981
