VinUni Banner

Advancing Humanoid Robot Learning through Perception Modeling, World Model, and Biomimetic Machine Intelligence

Advancing Humanoid Robot Learning through Perception Modeling, World Model, and Biomimetic Machine Intelligence

Advancing Humanoid Robot Learning through Perception Modeling, World Model, and Biomimetic Machine Intelligence

Successful applicants will join the College of Engineering and Computer Science (CECS)VinUniversity as PhD Students, Master’s Researchers, or Research Assistants, working with VinUni’s robotics research group and international partner labs. 

Project Information 

This project aims to develop advanced robot learning algorithms for humanoid robots, enhancing their ability to perceive, model the world, and act with greater autonomy, safety, and long-horizon capability.

The team focuses on biomimetic perception models, improving safety-aware behaviors, strengthening long-term learning, and enabling adaptation in real-world environments.

Key research approaches include perception modeling, world modeling, imitation learning, reinforcement learning and large-scale behavioral models.

The project is led by Dr. Dung D. Le (VinUniversity) in collaboration with an international team of experienced researchers: Viet Dung Nguyen, Rochester Institute of Technology (USA); Dr. Minh Nhat Vu, Vienna University of Technology – TU Wien (Austria); Dr. An Thai Le, VinUniversity / VinRobotics and Dr. Quang Anh Nguyen, University of Liverpool (United Kingdom)

Participants will have the opportunity to develop algorithms, build experimental systems, and contribute to the team’s research deliverables and outcomes.

Research Objectives 

The project focuses on the following directions:

– Developing robot learning algorithms that improve long-horizon task performance, stability, and safety of humanoid robots.

– Building biomimetic perception and action models to enhance generative modeling and contextual understanding.

– Implementing learning methods such as imitation learning and reinforcement learning to optimize robot skills and behaviors.

– Increasing autonomous exploratory capabilities and reducing dependence on expert demonstrations.

– Integrating learning systems into real-world environments and online-learning settings for real-time adaptation.

– Collecting data, developing benchmarks, and building behavioral models for both simulation and real-world scenarios.

Project Contact 

For details, please contact Dr. Dung D. Le via email [email protected]

Graduate Admissions Contact 

Learn more about Master’s, PhD programs, and scholarships at: VinUni Graduate Research Excellence Program

Banner footer