Thesis Proposal - Data Driven Diagnostics and Profiling in Autonomous Driving development system

Background: 

Autodrive is an autonomous driving development system designed for prototyping and testing control algorithms, path planning, vehicle sensor layout, etc. The system consists of a physics engine (Unreal Engine), a middleware (ROS), CARLA (signal/sensor interface) and a control and decision software developed by Syntronic. Autodrive can be used for simulations or deployed on physical hardware (PiCar).

The system uses multiple methods of communication between the different layers, including publisher-subscriber networks for signal handling, I2C and TCP/IP for PiCar, USB for physical LIDAR, etc. Developing a model and/or tools for diagnosing and profiling of Autodrive communication will be useful in continued development and scaling of the system. Such a model could also be used to identify and analyze problems with Autodrive, such as bottlenecks and unnecessary signal/data load.

Work description: 

The work will consist of collecting relevant data, identifying relevant issues, and designing and developing tools and/or models for diagnosing and profiling Autodrive communication over the various networks.

We think the thesis will contain the following parts,

• Collect data to analyze
• Developing and test a model or tool for system diagnostics
• Compare the system before and after the model was developed and/or deployed

For a successful thesis work, we believe that the applicants should have a MSc. in Applied Physics, Computer Science, or other equivalent program. 

Application:

We look forward to receiving your resume, and preferably, a personal letter in which you explain why you want to write your thesis with Syntronic.

We screen and evaluate applications on an ongoing basis.