Master's thesis; AI Architects for the Next Generation of Autonomous Robots

Background

The next leap in autonomous robotics is moving beyond simple obstacle avoidance towards genuine environmental understanding. For a robot like a lawn mower to operate intelligently and safely in complex human environments, it needs to understand context—distinguishing between a harmless patch of leaves and a child’s forgotten toy, or identifying a newly planted flowerbed to avoid. In collaboration with Husqvarna, the global leader in robotic lawn mowers, RISE is launching an ambitious project to build the advanced AI brains required for this next generation of smart, autonomous systems.

Project Description

This Master's thesis project is for two students who will work as a team to design, build, and test the AI architecture for a future Automower. The goal is to create a robust and adaptive intelligence that enables the robot to perceive, reason, and act with unprecedented sophistication.

Together, you will explore and compare different state-of-the-art AI paradigms to find the optimal solution. The project will involve investigating several exciting research paths:

• Modular, On-Device Intelligence: Designing multi-agent systems where a Vision-Language Model (VLM) acts as the "eyes" and a Small Language Model (SLM) serves as the "brain" for efficient, on-board decision-making.

• End-to-End On-Device Agents: Exploring the latest compact Vision-Language Agents (VLAs)—powerful, single models that can map pixels directly to actions and are efficient enough to run on embedded hardware.

• Cloud-Powered Reasoning: Leveraging the power of large-scale VLAs running in the cloud, potentially in a future 6G context, to provide a "gold standard" of behaviour and to explore hybrid robot-cloud architectures.

The two students will collaborate closely, with the flexibility to specialise in different aspects of these research paths based on their interests and the project's evolution.

Key Responsibilities

• Design and implement novel AI architectures for robotic control.

• Train, fine-tune, and evaluate deep learning models (VLMs, SLMs, VLAs) for robotic tasks.

• Deploy and test your solutions on a physical robotic platform.

• Rigorously analyse and benchmark the performance, efficiency, and safety of different approaches.

• Collaborate with senior researchers at RISE and experts from Husqvarna.

Qualifications

• Ongoing Master’s degree in Computer Science, AI, Robotics, or a related field.

• A strong background in AI, deep learning, and computer vision.

• Proficiency in Python and experience with ML frameworks like PyTorch or TensorFlow.

• A keen interest in robotics, autonomous systems, and AI agent architectures.

• Direct experience with VLAs is not required; a strong foundation in deep learning and a passion for robotics are what matter most.

• A creative, analytical, and collaborative mindset.

What we offer

• A chance to work on a scientifically novel project that is defining the future of AI in robotics.

• Direct collaboration with Husqvarna, a world leader in robotics, shaping their future products.

• Access to RISE’s advanced resources, unique datasets, and test environments.

• Opportunities for professional growth and networking, including potential PhD opportunities.

• A friendly and dynamic research environment with experienced supervisors.

Terms

• Scope: The thesis comprises 30 credits (hp/ECTS) per student.

• Start Date: Spring 2026, or by agreement.

• Location: RISE, Kista, Stockholm, with the possibility for some remote work.

• Compensation: In line with RISE guidelines for strategically important projects, a compensation of 30,000 SEK per student is offered upon completion and approval of the 30-credit thesis.

• Please note that due to industrial confidentiality and non-disclosure agreements, this position is restricted to EU citizens only.

Welcome with your application!

Applications must be submitted through the our recruitment portal Teamtailor.

Last day of application: 27th of October, 2025.

For more information, please contact:

• Dr Fehmi Ben Abdesslem, fehmi.ben.abdesslem@ri.se

• Dr Joakim Eriksson, joakim.eriksson@ri.se

• Miriana Passarotto, miriana.passarotto@ri.se