Thesis - Edge-Deployed Physics-Aware Vehicle Dynamics Modelling
Join this exciting thesis project to develop a hybrid model combining vehicle dynamics with neural networks for accurate, real-time state predictions, optimized for edge devices. Apply now!
We are Zeekr Technology Europe. Powered by engineers and creative minds from all over the world. Located in Gothenburg, Sweden, one of the leading mobility innovation grounds worldwide. Established in 2013 as CEVT and began with the mission to create a revolutionary modular vehicle architecture, CMA. As of 2024, CEVT is now Zeekr Technology Europe, signaling our belonging within Zeekr. Delivering intelligent, sustainable, and user centric technology and software for tomorrow's electric mobility today.
Project Description:
The prediction of vehicle dynamics is crucial for ensuring safety and performance in various automotive applications. Traditional physics-based models offer high accuracy but are often cumbersome to implement and tune, requiring extensive testing setups. Conversely, purely data-driven approaches like neural networks provide flexibility but lack generalization and adherence to physical constraints. This dichotomy underscores the need for a hybrid approach that leverages the strengths of both methodologies, aligning with the principles of Physics-Informed Learning (PIL) and Physics-Augmented Learning (PAL).
Scope
This research aims to develop a novel method that integrates physics coefficient estimation and dynamical equations with neural networks to predict vehicle states accurately under varying speed conditions. The method will incorporate a physical feasibility layer to ensure that internal coefficient estimates remain within their nominal physical ranges, thus maintaining the reliability and accuracy of predictions. The vehicle model will be designed to be generic enough to perform lateral and longitudinal state estimation with minimal transfer learning. Furthermore, the method will be optimized for deployment on edge devices such as microcontrollers, ensuring real-time performance and low computational overhead.
Objectives
- Develop a Hybrid Model: Combine physics-based vehicle dynamics equations with neural network capabilities to leverage the strengths of both methodologies, in line with PIL and PAL frameworks
- Implement Physical Feasibility Layer: Ensure that coefficient estimates adhere to physical constraints, enhancing the model's reliability and accuracy.
- Validation and Comparison: Validate the model against standard vehicle dynamics tests and compare its performance to existing methods to demonstrate its efficacy.
- Versatility and Scalability: Ensure the model's applicability for both high-speed and low-speed conditions, with effective lateral and longitudinal state estimation, and minimal need for transfer learning.
- Optimization for Edge Devices: Optimize the model for deployment on edge devices, ensuring real-time performance and low computational overhead.
Expected Outcomes
- Robust and Accurate Model: Achieve a vehicle state prediction model that combines the advantages of physics-based and neural network approaches, offering robust and accurate predictions.
- Reliable Predictions: Demonstrate the model's capacity to provide reliable predictions under varying speed conditions, aligning with the principles of PIL and PAL.
- Scalable Methodology: Establish a scalable methodology that can be adapted for various vehicle dynamics applications, with minimal transfer learning required for lateral and longitudinal state estimation.
- Real-Time Performance: Deploy the model on edge devices, showcasing its real-time performance and low computational overhead.
Your skills and background
- Master’s degree in Computer Science, Vehicle Engineering, Physics, or relevant fields
- Strong background in Control Systems, Vehicle Dynamics, and Vehicle Modeling
- Strong programming skills in MATLAB, Python, C++, or similar languages
- Strong background in Machine Learning and Data Science, with experience in working with time-series data
- Experience with Machine Learning frameworks (e.g., TensorFlow, PyTorch)
- Familiarity with Embedded Systems and Edge Computing, including experience with microcontroller programming and deployment.
- Strong analytical and problem-solving skills
Why you should join Zeekr Tech Eu
We are engineers, developers, and innovators from around the world. Joined together by entrepreneurship, our unique blend of global culture, and a belief in a smarter more sustainable future. At Zeekr Tech Eu we fast-track innovation and transform ideas into pioneering technology solutions, doing your master thesis here is no different. We are convinced that a thesis project is a major contribution to our innovation capabilities and long-term development. You'll have a great opportunity to use your skills and creativity to push the boundaries of what´s possible.
What happens when you apply
If this sounds interesting and you match the requirements, please don't hesitate to submit your application with a CV and cover letter. Shortlisted candidates will be contacted for an interview to further discuss the project's details and expectations.
This thesis project is intended for 2 students
Don't hesitate to get in touch with the supervisors for more information about the project:
- Karthik Prasad, karthik.prasad@zeekrtech.eu
- Utsav Khan, utsav.khan@zeekrtech.eu
Starting date: January 2024
Last application date: 2024-11-06
Apply today. We will perform ongoing selection during the application period. We look forward to hearing from you!
Please note that due to GDPR regulations, we can only accept applications sent through the recruitment system, not via email or other channels.
- Department
- Student
- Role
- Master Thesis Student
- Locations
- Göteborg
Göteborg
Some extra fuel for our co-drivers
-
Vacation
30 days of paid vacation and additional 8 days paid ATK-days annually for full time employees. The ATK-days are set by the company, and often placed in connection to the national holidays.
-
Short Term Incentive (STI)
We want to reward your performance! You are therefore eligible to our annual STI program. The program is designed with one part connected to the business result, since we are a team, and another part connected to your individual achievements.
-
Flexible workplace and a modern office space in great location
At Zeekr we believe that where you do your work is best determined and agreed upon within each team – among you, your team, and your manager.
Our bright and modern office, Uni3 by Geely, is located at Lindholmen, close to the city center of Gothenburg, a vibrant innovation hub and technology cluster. -
Of course...
...we do have a Collective Agreement with the unions. For instance, pensions and insurances are regulated by Teknikavtalet. We also have Wellness allowance, Benefit portal and favorable private leasing of cars from the Geely family.
About Zeekr Technology Europe
We are Zeekr Technology Europe. Powered by engineers and creative minds from all over the world. Located in Gothenburg, Sweden, one of the leading mobility innovation grounds worldwide. Delivering intelligent, sustainable, and user centric technology and software for tomorrow's electric mobility today
Thesis - Edge-Deployed Physics-Aware Vehicle Dynamics Modelling
Join this exciting thesis project to develop a hybrid model combining vehicle dynamics with neural networks for accurate, real-time state predictions, optimized for edge devices. Apply now!
Loading application form
Already working at Zeekr Technology Europe?
Let’s recruit together and find your next colleague.