OR0317: Internship - Vision-Language-Action (VLA) Models for Robotic Manipulation MERL is seeking a highly motivated Ph.D. student to conduct research on Vision-Language-Action (VLA) models for robotic manipulation. Recent VLA models have demonstrated impressive capabilities in general-purpose robotic manipulation. This internship will investigate new methods for improving the reliability and effectiveness of VLA policies in real-world manipulation tasks. Potential research directions include exploring VLA-based approaches, multi-arm coordination, and adaptive behaviors for industrial manipulation tasks, with particular emphasis on reliability assessment, policy intervention, and corrective-action strategies. The selected intern will work closely with MERL researchers to develop and implement novel algorithms, deploy and evaluate learning-based policies on robotic platforms, and conduct both simulation and real-world experiments. The intern will also disseminate research findings at leading robotics conferences, including ICRA, RSS, CoRL, and IROS. The internship start date and duration are flexible. Interested applicants are encouraged to apply with an updated CV and a list of relevant publications. Required Specific Experience Current enrollment in a Ph.D. program Strong background in robot learning for manipulation Hands-on experience with Vision-Language-Action (VLA) models for robotics, such as pi0, OpenVLA, Octo, GR00T N1, or related state-of-the-art models Proficiency with ROS, Python, and deep learning frameworks such as PyTorch Experience with robotic simulation environments and sim-to-real transfer Experience deploying and evaluating learning-based policies on physical robotic platforms A strong publication record or demonstrated research potential The pay range for this internship position will be 6-8K per month.

Research Areas:

Robotics, Computer Vision, Artificial Intelligence, Machine Learning

Host:

Siddarth Jain APPLY NOW