Thesis work - AI-based human sensing using WI-FI signals

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 is continuously exploring new technologies that can support safer, more efficient, and human-centred production environments. Within the Smart Factory Lab, emerging technologies are investigated and evaluated before their potential application in industrial settings.

 Understanding human movement and posture is valuable in several industrial applications, including ergonomics, workplace safety, and human–machine interaction. Today, such information is commonly obtained using cameras, wearable sensors, or motion-capture systems. While these technologies can provide detailed information, they may also introduce limitations related to privacy, installation, visibility, or the need for workers to wear additional equipment.

 Recent advances in wireless sensing show that Wi-Fi signals can potentially be used to detect human presence, movement, posture, and other activities. As Wi-Fi signals propagate through an environment, interactions with the human body cause measurable changes in the wireless channel. These changes can be captured through Channel State Information (CSI) and analysed using signal-processing and machine-learning methods. This creates an interesting opportunity to investigate whether existing Wi-Fi infrastructure and low-cost sensing hardware could provide a non-contact and privacy-preserving approach to understanding human movement in industrial environments.

Objective

The main goal of this thesis is to explore the feasibility of using Wi-Fi Channel State Information (CSI) to sense human posture and movement without relying on cameras or wearable sensors. Wi-Fi signals interact with the human body through reflection, attenuation, and multipath propagation. These variations can be captured as CSI and potentially used to infer information such as presence, movement, posture, body orientation, and, at a more advanced level, human pose.

Job Description

The thesis will follow a structured research process, beginning with problem formulation and requirement elicitation in collaboration with domain experts. This will be followed by a systematic review of relevant literature to establish the state of the art, identify research gaps, and select appropriate methods, datasets, and evaluation metrics. Based on these findings, a well-defined experimental methodology will be developed, including the study design, data collection strategy, implementation approach, and evaluation protocol. The selected methods will then be implemented and empirically evaluated through controlled experiments. The results will be analyzed with respect to accuracy, robustness, limitations, and practical applicability. Finally, the methodology, findings, conclusions, and recommendations will be documented in the thesis and communicated through a final presentation.

Education/program/focus 

Master's program in computer science, embedded systems, artificial intelligence, machine learning, electrical engineering, signal processing, robotics, mechatronics, or a related technical field.

Number of students: 1

Start date: January 2026

Estimated time needed: 20 weeks

Contact persons and supervisors 

Veeresh Elango (veeresh.elango@scania.com)

Marian van Helmond (marian.van.helmond@scania.com)

Application

Your application must include a CV, personal letter and transcript of grades

• Optional: propose a tentative approach to the problem

Publication date from-to.

2026-10-01–2026-11-22