Vacant job

Lund University was founded in 1666 and is consistently ranked among the world's leading universities. We have approximately 46,000 students and 8,500 employees in Lund, Helsingborg, and Malmö. We are united in our strive to understand, explain, and improve our world and people's conditions.

Description of the activity

The doctoral student will be part of the Robotics and Semantic Systems (RSS) division, which is part of the Department of Computer Science. RSS consists of several closely collaborating research groups conducting research and education in robotics and AI, intelligent autonomous systems, robot learning, machine learning, human-robot interaction, and language technology.

RSS, together with the Department of Automatic Control, operates RobotLab LTH, a research and educational environment equipped with a large number of robot-based and related systems, ranging from small drones and UAVs to classic industrial manipulators. The research environment conducts internationally recognized research in robot learning, embodied AI, robot manipulation, semantic scene understanding, and cognitive robotics. The Robot Lab combines machine learning, computer vision, robot control, long-term planning for real-world robot systems, and physical AI.

The doctoral student will work in close collaboration with other researchers and interact with cross-disciplinary teams within robotics, machine learning, perception, and AI reasoning. The Robot Lab provides access to advanced robot platforms, including a dual-arm KUKA iiwa and UR5e manipulators, ROS 2-based robot infrastructures, simulation environments, and modern GPU computing resources.

Being a doctoral student

As a doctoral student, you are both admitted as a student and employed at Lund University.

As a doctoral candidate, you are trained in a scientific approach. This can be briefly described as gaining practice in critical and analytical thinking, solving problems independently using appropriate methods, and developing an awareness of research ethics. Additionally, as a doctoral student, you are given the opportunity to work on projects, develop your leadership skills, and enhance your pedagogical abilities. Throughout your studies, you are guided by supervisors. The doctoral studies conclude with a thesis and a doctoral degree.

More about being a doctoral student at LTH can be found at lth.se.

Subject and project description

The doctoral appointment is within computer science with a focus on embodied AI, robot learning, and reliable Vision-Language-Action (VLA) systems. The project is part of ELLIIT's research initiative “A Robust and Reliable Vision-Language-Action Interface”.

Recent advances in Large Language Models (LLM), Large Reasoning Models (LRM), and Vision-Language-Action (VLA) systems have enabled robots to perform increasingly complex tasks based on multimodal sensory input and natural language instructions. However, current systems still lack robust situational awareness, introspection, uncertainty estimation, and reliable ability to plan over long time horizons.

The doctoral project focuses on developing reliable robot systems that combine:

• symbolic and semantic world models.
• large reasoning models for planning over long time horizons.
• Vision-Language-Action models for robot execution.
• detection of out-of-distribution data (OOD) and uncertainty estimation.
• adaptive robot behavior based on predicted probabilities of task success.

A central research challenge is to develop robot systems that can reason about whether a planned action is likely to succeed in a given situation and adjust execution parameters accordingly, for example, by reducing speed or choosing alternative actions in uncertain situations.

The project will investigate how symbolic world models, semantic scene representations, and uncertainty-aware reasoning can be integrated into embodied AI systems to improve robustness, explainability, and safety in real-world robot applications.

Job responsibilities

You will primarily focus on your doctoral education, which includes participating in research projects, doctoral courses, seminars, and conferences.

The responsibilities include:

• Conducting research in embodied AI, robot learning, and uncertainty-aware robot reasoning,
• Developing symbolic and semantic world models for robot planning,
• Integrating Large Reasoning Models (LRM) with Vision-Language-Action (VLA) systems,
• Developing methods for out-of-distribution detection (OOD) and uncertainty estimation in robot-based decision-making,
• Implementing and evaluating robot systems on real robot platforms,
• Publishing research results at leading international conferences and in scientific journals,
• Collaborating with researchers and industrial partners in robotics and AI,
• Contributing to the maintenance and development of robotics research infrastructure.

The job responsibilities also include participation in teaching and other departmental activities, up to a maximum of 20 percent of working time.

Qualifications

To be admitted and employed as a doctoral student, you must meet the requirements below.

For the full advertisement, please see: https://lu.varbi.com/what:job/jobID:937557/ (https://lu.varbi.com/what:job/jobID:937557/)

We welcome your application!

LTH – Lund University Faculty of Engineering – is part of Lund University. At LTH, we educate people, build knowledge for the future, and work hard to develop society. We create space for brilliant research and inspire the creative development of technology, architecture, and design. Nearly 12,000 students study here. Each year, our researchers – many of whom work in world-leading profile areas – publish approximately 100 dissertations and 2,000 scientific findings. A number of research results and student projects are refined into innovations. Together, we explore and create – for the benefit of the world.

We decline all contact from advertising sales, recruitment, and staffing agencies due to public procurement regulations.