Thesis Student - Decentralized Fault Handling for Vehicle Localization in Multi-Sensor Localization

30 credits - Decentralized Fault Handling for Vehicle Localization in Multi-Sensor Localization

Introduction

A thesis project at TRATON provides a unique opportunity to gain practical experience, expand your professional network, and contribute to the future of sustainable transportation. Many of our current employees began their careers at Traton by completing their degree projects within the organization.

TRATON Group includes leading commercial vehicle brands such as Scania, MAN, Volkswagen Truck & Bus, and International. Together, we are committed to driving the transformation toward sustainable transport solutions. As a thesis student, you will work closely with experienced engineers and specialists, contributing to real-world product development while gaining valuable industry experience.

For more information, visit www.traton.com.

Background

Robust localization is important for future transport systems and from a SOTIF and ODD perspective, localization must also handle corner cases such as GNSS disturbance, urban canyons, tunnels, indoor or depot operation, and situations with few visible satellites.

Integrated navigation combines global localization sources, such as GNSS/RTK and perception-based localization, with dead-reckoning sensors such as IMUs, wheel-speed sensors and radar odometry. These sources have different strengths and weaknesses, which makes robust availability assessment, fault detection and graceful degradation important across the vehicle ODD.

TRATON needs practical methods for detecting, diagnosing and isolating localization errors. A decentralized architecture with local filter modules for different sensor groups can keep sensor paths clearer and make fault handling easier, while a centralized filter can continue to provide the best navigation performance.

Assignment

Phase 1: Establish a shared stochastic filtering and validation foundation for 3D navigation using IMU, vehicle odometry and baseline GNSS. MATLAB, Python and C++ are possible tools. Python/Databricks is likely preferable, because ground truth is available in Databricks in the cloud for evaluating fusion errors.

Phase 2: The work should start with a literature and market review, carried out together with the supervisors, to scout for additional localization sources that could complement GNSS and perception-based localization. Development and testing of methods for error detection, diagnosis and isolation in localization, adapted to the student profile and the available measurements. If real measurements are insufficient, the POC may use synthetic data and virtual measurements with injected known errors. Innovation-based methods such as GLR may be evaluated for selected fault types, together with other specialized fault detectors where relevant.

Additional comments: Phase 1 should preferably be done together with the companion thesis student. The students establish a common foundation and then split into their Phase 2 tracks. The methods should support autonomous driving and ADAS Next Generation functions requiring robust localization, redundancy reasoning, availability assessment and graceful degradation.

Education / Program / Qualifications

We are looking for a curious, motivated, and hands-on student with a strong interest in embedded systems, multiple sensor technologies, electrical architecture, and sustainable transportation.

Preferred Qualifications

Interest in sensor technology, embedded systems and automated driving systems.

Fluency in English, both written and spoken.

Education:

Students from engineering with emphasis in robotics, signal processing, control, automation, autonomous systems, computer science, electrical engineering, mechatronics, mathematics, applied physics or similar.

Thesis Start Date

• February 2027

The exact start date and project timeline will be agreed upon between TRATON, the selected candidate, and the academic institution.

Contact Information

For questions regarding the thesis project, please contact:

• Bengt Boberg, Senior Development Engineer - TGRAAP1

bengt.boberg@scania.com

• Laís Ramos, Manager - TGRAAP1

lais.carolina.ramos@scania.com

Application

Your application must include:

• CV
• Personal letter
• Transcript of grades

We welcome applicants from all backgrounds and believe that diverse perspectives contribute to better innovation, stronger teams, and improved solutions.

Please note that a background check may be conducted for this position. Interviews will be held continuously, and the recruitment process may close before the stated application deadline.

Join us in shaping the future of battery technology and sustainable transportation. We look forward to receiving your application.

Requisition ID: 34689

Number of Openings: 1.0

Part-time / Full-time: Full-time

Permanent / Temporary: Temporary

Country / Region: SE

Location(s):
Södertälje, SE, 151 38

Required Travel: 0%

Workplace: Hybrid