Thesis Work - Automation utilizing Computer Vision
At a glance
Mid-level AI Research role at Scania. Södertälje.
Pay not stated AI Research salary
aimlroles summary, based on the employer's posting.
What you'll do
- Combine information from several camera views and test whether it improves object detection and verification over one camera
- Build a working proof of concept for a relevant production or logistics scenario using Scania’s standard vision software
- Assess how camera placement and spatial relationships affect detection reliability and feature checks
- Compare multi-camera and single-camera methods for robustness, accuracy, and object verification
What you bring
- You are studying in a master’s program in machine or production engineering, computer science, data science, or a related field
- You can investigate computer vision methods and develop a proof of concept for an industrial use case
Who this fits
This thesis suits a master’s student interested in computer vision for production or logistics. It is based in Södertälje, Sweden, runs for an estimated twenty weeks, and is due to start in January 2027. The project is for one student; your application must include a CV, personal letter, and grade transcript.
From the employer
30 hp - Smart Factory Lab
Introduction
Thesis work is an excellent way to get closer to Scania and build relationships for the future. Many of today's employees began their Scania career with their degree project.
Background
Scania has recently standardized its approach to computer vision in production and logistics by providing a global software tool. This supports the company’s goals of increasing automation, building in-house knowledge, and advancing digital transformation.
Previous work has explored how this technology can be applied within logistics, including the use of computer vision for detecting and reading information on incoming material and boxes. Building on this work, there is a need to further investigate how computer vision can be used in more complex scenarios where objects need to be detected and verified from multiple viewpoints.
In many industrial applications, a single camera may not provide enough information to reliably detect or verify all relevant features of a three-dimensional object. Occlusion, viewing angle, object orientation, and environmental conditions can all affect the robustness of the solution.
Objective
The work will investigate how information from multiple camera views can be combined to improve the reliability of object detection and verification compared with using a single camera. This may include reasoning about the spatial relationship between cameras, objects, and detected features.
The solution will be developed using Scania’s standardized computer vision tool, ensuring that the work contributes to the continued development of scalable and reusable computer vision solutions within Scania.
The goal is to explore best practices for multi-camera computer vision, develop a working proof of concept in a relevant industrial use case, and evaluate how multi-camera spatial reasoning can improve robustness, detection accuracy, and verification of objects and their features.
Key questions (Inspiration)
- How can information from multiple cameras be combined to improve detection and verification of data and spatial objects relationships?
- How can Scania’s standardized computer vision tool be applied, adapted, and integrated to support multi-camera solutions and ensure scalability across different production and logistics applications?
- What are the advantages and limitations of a multi-camera approach compared with a single-camera solution?
Education/program/focus
Master's program in machine engineering, production engineering, management, computer science, data science, or related fields.
Number of students: 1
Start date: January 2027
Estimated time needed: 20 weeks
Contact persons and supervisors
Robin Törnblom, robin.tornblom@scania.com
Application:
Your application must include a CV, personal letter and transcript of grades
A background check might be conducted for this position. We are conducting
interviews continuously and may close the recruitment earlier than the date
specified.
Publication date from-to.
2026-10-01–2026-11-22