Master thesis Image-based path planning and autonomous drone control along conveyors

The AkuTra project is developing an intelligent system for monitoring the condition of belt conveyor systems. It comprises a measuring unit and a transport unit, which is implemented by a drone (UAV). The master's thesis aims to develop image-based navigation and path planning so that the drone can fly autonomously and transport the measuring unit to defined measuring stations. Relevant system structures are detected from the camera feed using object recognition (e.g., YOLOv4) and integrated into a ROS-based control system. The method is modeled and validated in a simulation environment such as Gazebo.

Your tasks

As part of this master's thesis, image-based navigation and ROS-based control for the transport system will be developed and tested in simulations. Your tasks in detail:

• Research and familiarization with image-based navigation and path planning methods and autonomous drone control
• Creation of a concept for implementation
• Software implementation of image-based navigation and control
• Simulation validation of the system
• Critical evaluation of the results
• Documentation

Your profile

You are studying one of the following subjects:

• Computer science
• Data science
• Mechanical engineering
• Mechatronics
• Electrical engineering
• or similar

It would be ideal if you also:

• have good written and spoken German skills,
• have good programming skills in Python and data analysis (e.g., NumPy, pandas, matplotlib) and
• have experience with ROS/ROS2 and image processing.

Are you interested in data analysis, drones, and artificial intelligence? And do you enjoy familiarizing yourself with new topics independently? Then we look forward to receiving your application!

We offer

• Independent work
• Flexible working hours
• Well-equipped workplaces
• Home office by arrangement
• Test implementation
• Long-term collaboration, if applicable