Master Thesis, 30 HP: Orientation Estimation

Are you a student eager to turn your theoretical expertise into real-world impact? At Saab, we believe innovation accelerates when fresh perspectives meet critical missions. Your master thesis could power our next leap in robust orientation estimation for training and operational systems.

You will explore how modern computer vision and sensor fusion can deliver accurate, stable, and trustworthy orientation (roll, pitch, yaw) under demanding conditions. Working alongside experienced engineers and researchers, you will prototype algorithms, evaluate them on realistic datasets, and translate insights into practical recommendations for deployment in training ranges and onboard platforms.

Background

Saab AB, Training and Simulation, delivers realistic ground combat training systems, from brigade-level exercises to individual soldier training. Reliable orientation is foundational for vehicle state awareness, gunnery stabilization, and shooter performance evaluation. By combining cameras with inertial measurement units (IMUs), visual–inertial approaches (e.g., ORB‑SLAM3 as a reference) can provide resilient orientation tracking even when GPS is unavailable, environments are dynamic, or motion is fast. In this thesis, the focus is on orientation estimation; position/odometry is not the primary goal, though camera-based motion estimates may be explored for analysis.

We aim to advance orientation estimation using multisensory modules—e.g., stereo or monocular cameras tightly coupled with IMUs—and to understand how sensor choice, placement, and calibration affect accuracy, latency, and robustness. The outcome will inform next‑generation training analytics and operational systems where dependable orientation is mission‑critical.

Description of the master thesis

Selected candidates will design, implement, and evaluate computer vision and sensor fusion approaches for visual–inertial orientation estimation and sensor optimization.

You will:

• Integrate and adapt orientation-focused visual–inertial pipelines (e.g., ORB‑SLAM3 as a reference) with IMUs for real-time attitude and heading

• Investigate minimal yet effective sensor configurations (mono vs. stereo, FOV, frame rate, IMU grade) and placements

• Develop interpretable metrics and feedback relevant to marksmanship and vehicle dynamics

• Provide confidence indicators on orientation estimates for trustworthy training feedback

• Benchmark on representative datasets and, where possible, in lab/range settings

Example 1 – Camera + IMU Orientation Estimation (ORB‑SLAM3‑inspired) Build a camera+IMU pipeline that prioritizes orientation accuracy and stability under motion blur, low texture, and lighting variations. Use ORB‑SLAM3 (or similar) as a conceptual and benchmarking reference, while implementing the thesis pipeline in Python or MATLAB. Key elements include:

• Camera–IMU time synchronization and extrinsic calibration

• IMU bias handling and drift mitigation

• Visual orientation constraints from multi-view geometry (e.g., essential matrix for relative rotation, robust feature tracking)

• Sliding‑window estimation targeting orientation states only

• Calibrated uncertainty suitable for real-time training feedback Optional: derive camera-based motion estimates for analysis, but keep the primary focus on orientation, not full odometry.

Example 2 – IMU + Complementary Sensor for Robust Heading Fuse an IMU with a complementary sensor to strengthen yaw/heading, especially where visual observability is weak. Candidate modalities (choose based on environment and availability):

• Magnetometer with hard/soft‑iron compensation and disturbance detection

• LiDAR-based scan matching for relative yaw stabilization without relying on global position

• Automotive FMCW radar (ego-rotation via scan matching or Doppler constraints) to stabilize heading Emphasize robustness to disturbances and provide confidence indicators for training feedback.

Applications will be reviewed on a rolling basis and positions may be filled before the application deadline.

We seek students completing master's studies in Applied Physics and Electrical Engineering, Systems Control and Mechatronics, or Complex Adaptive Systems (or similar) with strong competence in:

• Computer vision and multi-view geometry

• Signal processing and state estimation / sensor fusion

It is a plus if you have:

• Prototyping experience in Python or MATLAB (OpenCV, NumPy/SciPy, MATLAB Computer Vision and Sensor Fusion Toolboxes)

• Experience with camera–IMU calibration (e.g., Kalibr), time synchronization, and sensor placement

• Familiarity with ORB‑SLAM3 concepts and datasets as references for orientation evaluation

• Background in embedded systems constraints and real-time algorithm design

• Interest in interpretable metrics for training feedback and human–machine interaction

Please include your university grades with your application.

We provide the mentorship, tools, and test environments to help you transform theory into field-ready capability. Join us to push the boundaries of robust orientation estimation for training and beyond.

This position requires passing a security vetting process in accordance with current regulations on security protection. For roles requiring security clearance, additional obligations on citizenship may apply.

Explore a wealth of possibilities. Take on challenges, create smart inventions, and grow beyond. This is a place for curious minds, brave pioneers, and everyone in between. Together, we achieve the extraordinary, each bringing our unique perspectives. Your part matters.

Saab is a leading defence and security company with an enduring mission, to help nations keep their people and society safe. Empowered by its 26,100 talented people, Saab constantly pushes the boundaries of technology to create a safer and more sustainable world.

Saab designs, manufactures and maintains advanced systems in aeronautics, weapons, command and control, sensors and underwater systems. Saab is headquartered in Sweden. It has major operations all over the world and is part of the domestic defence capability of several nations. Read more about us here.

Last application day

Contact information

Oscar Gunnarsson