Internship - Perception and Control for Legged Loco-Manipulation

Job Description

Job Description

MERL is seeking highly-qualified Ph.D. interns passionate about robotic research to contribute to the fundamental research of perception and control algorithms for loco-manipulation tasks. The work will involve extensive validation in both simulation and hardware, using a legged robot equipped with a manipulator arm with mounted cameras. Depending on the intern's expertise and the agreed project scope, research topics may include one or more of the following: Human-robot collaboration, LLM/VLM-based Active SLAM, and state estimation. The internship is expected to lead to a publication in a top-tier robotics, machine learning, or computer vision conference and/or journal. The internship has a flexible start date (Spring/Summer 2026), with a duration of 3-6 months.

Required Specific Experience

• Current/Past Enrollment in a Ph.D. Program in Computer Engineering, Computer Science, Electrical Engineering, Mechanical Engineering, or a related field.
• Exposure to core subjects from robotics curriculum including planning, control, and estimation (e.g., motion planning, feedback control, sensor fusion, Kalman filtering, kinematics, dynamics, etc).
• Expertise in at least one or more areas: LLM fine-tuning, SLAM, graph-based optimization (e.g, factor graphs with GTSAM), generative models (e.g., diffusion, flow-matching), 3D computer vision, whole-body control.
• Excellent programming skills in Python and/or C++, with prior knowledge of ROS2 and high-fidelity simulators such as Isaac Sim and/or MuJoCo.

The pay range for this internship position will be 6-8K per month.