Autonomous Road Surveying

Fig. 1: Track in the Lab im Labor Autonome Systeme

Autor:	Qamar Sajjad
Art:	Bachelorarbeit/bachelor thesis
Starttermin:	TBD
Abgabetermin:	TBD
Betreuer:	Prof. Dr.-Ing. Schneider

Introduction

A model car (scale 1:20) equipped with a camera should drive autonomously in the right lane. Usually this is done by image processing. In this bachelor thesis steering angle and video are used to train an artificial intelligence (AI) to drive autonomously on the track (see Fig. 1)

Task list

1. Familiarization with the topic
2. Set up requirements for the system
3. Research on solutions for the task
4. Implementation in MATLAB^®. Teach a AI to drive in the right lane. Input are video and steering angle
5. Evaluation of the solutions using a morphological box ([https://en.wikipedia.org/wiki/Morphological_box Zwicky box)
6. Implementation the most promising on the JetRacer
7. CAD design and 3D printing of prism mount for the JetRacer
8. Prism assembly and recording of the vehicle movement
9. Evaluate the results based on technical features
10. Discussion of the results
11. Testing of the system requirements - proof of functionality
12. Scientific documentation as a wiki article with an animated gif

Knowledge Requirements

The project requires prior knowledge in some but not all of the following subject areas. If you do not meet the requirements, the task can be individually adapted based on your previous knowledge.

• Model making (e. g. woodworking, metalworking, CAD, 3D printing)
• MATLAB^®/Simulink
• Robotics
• Control technology
• Document versioning with SVN
• Documentation with Word and in the HSHL Wiki.

Requirements for scientific work

• Scientific approach (project plan, etc.), useful article: Create Gantt chart
• Weekly progress reports (informative), update the table Meeting Minutes in conversation with Prof. Schneider
• Project presentation in the wiki
• Daily backup of work results in SVN
• Student work with Prof. Schneider
• Requirements for a scientific work

Getting started

• Wiki article: Lane Keeping with AI
• Deploying Deep Neural Networks to GPUs and CPUs Using MATLAB Coder and GPU Coder
• Deep Learning with MATLAB, NVIDIA Jetson, and ROS
• FAQ: Matlab to control jetracer(jetson nano tx1) motor
• Wiki article: JetRacer
• NVidia: JetRacer
• formulaedge.org
• Towards Autonomous Driving with Small-Scale Cars: A Survey of Recent Development
• Diese Library ist aus unserer Erfahrung ganz gut, um erst einmal den vielversprechendsten Algorithmus für die Test-Daten zu finden
• Schreiber, C.: KI-gestützte „Follow-Me“-Funktion am Beispiel des JetRacer. Mittweida, Hochschule Mittweida – University of Applied Sciences, Fakultät Ingenieurwissenschaften, Masterarbeit, 2023. URL: [1]
• Lane Detection: The 3 types of Deep Learning (non-OpenCV) algorithms
• Automatic Lane Line Detection using AI
• Lane Detection in Autonomous Vehicles Using AI
• Artificial Intelligence for Autonomous Vehicles : The Future of Driverless Technology.

Repository

• Video:
• Steering angle:

---

→ zurück zum Hauptartikel: Studentische Arbeiten