[PhD Position] Memory-Conditioned Vision-Language-Action Learning for Robotics

image_pdfimage_print

About the Position

We are looking for a highly motivated PhD student to join the Cyber-Physical Systems (CPS) Lab at Montanuniversität Leoben, Austria. The planned PhD topic is: Memory-Conditioned Vision-Language-Action Learning for Robotics. 

This research will investigate how memory mechanisms can be integrated into vision-language-action (VLA) models to enable robotic agents to retain, retrieve, and exploit past observations, interactions, and experiences for improved perception, reasoning, planning, and action. The potential research directions include, but are not limited to:

  • Episodic and semantic memory for embodied agents
  • Long-horizon vision-language-action learning
  • Memory-conditioned multimodal reasoning and decision-making
  • Open-vocabulary and language-conditioned robot perception
  • Continual and lifelong learning for autonomous robots
  • Spatial and temporal memory for embodied intelligence
  • Efficient deployment of multimodal foundation models on robotic systems
  • Evaluation of memory-augmented agents on real-world robotic tasks

The precise research topic is subject to change and can be adapted to the background, experience, and research interests of the successful applicant.

Candidate Profile

We are looking for applicants with a strong background in one or more of the following areas:

  • Robotics and autonomous systems
  • Machine learning and deep learning
  • Computer vision
  • Vision-language or multimodal learning
  • Reinforcement or imitation learning
  • Natural-language grounding
  • Robot perception and navigation

Experience with Python/C++, PyTorch, ROS/ROS 2, Isaac Sim, MuJoCo,  multimodal foundation models, 3D perception, or real robotic systems would be advantageous.

Applicants should hold, or be close to completing, a Master’s degree in Robotics, Computer Science, Artificial Intelligence, Electrical/Computer Engineering, Mechatronics, or a closely related discipline.

Position

  • Duration: 4 years
    Starting date: As soon as possible
  • Location: Chair of Cyber-Physical Systems, Montanuniversität Leoben, Austria
  • Employment type: Full-time (40 hours/week)
  • Salary: Salaried PhD position (€3,776.10/month), paid 14 times per year under the Austrian employment system
  • Application deadline: Open until a suitable candidate is found
  • Supervision: Prof. Elmar Rueckert and Dr. Nwankwo Linus

The successful candidate will work within an active robotics and machine-learning research environment, with access to robotic platforms and modern sensing systems, opportunities for international collaboration, and the expectation to publish in leading robotics and AI venues such as ICRA, IROS, RA-L, RSS, CoRL, NeurIPS, ICML, ICLR, and related conferences and journals.

How to Apply

Interested candidates should prepare the following application materials:

  • Curriculum Vitae
  • Academic transcripts
  • A short statement describing research interests, relevant experience, and motivation for the position
  • Letter(s) of recommendation, including the  referees’ contact information
  • Relevant publications, if available; otherwise, upload your MSc. thesis
  • Links to GitHub repositories, project pages, or other relevant research work.

Please upload all application materials using the following form: https://cloud.cps.unileoben.ac.at/index.php/apps/forms/s/DDwwb8yRJeCT5DoT67nXwaEd

In addition, please email the same application materials to: linus.nwankwo@unileoben.ac.at

Applications will be evaluated continuously, and the position will remain open until a suitable candidate is found. Please feel free to forward this announcement to interested students and colleagues. The Montanuniversität Leoben intends to increase the number of women on its faculty and therefore specifically invites applications from women. Among equally qualified applicants, women will receive preferential consideration.

Leave a Reply

Your email address will not be published. Required fields are marked *