The intern will research and implement reinforcement learning policies for robotic grasping and develop simulation environments for multi-finger robotic hands. They will also deploy these policies on hardware, analyze test data, and collaborate with the team on prototyping and documentation.
See what you’re missing
reinforcement learning
robot learning
control theory
Python
PyTorch
MuJoCo
Isaac Lab
PyBullet
CAD
mechanical design
Change the world. Love your job. This Systems Engineering Intern role is within Kilby Labs, the central R&D organization of Texas Instruments (TI). The lab is chartered to develop novel and innovative technologies and strategies for new business opportunities, drive next generation, differentiated technologies for existing businesses and recruit top technical talent for TI.
In this summer Systems Engineering intern role, you will explore AI-enabled control for robotic hands, with a focus on learning policies for reliable grasping and adaptive grip. You will train and evaluate policies in simulation, test them on robotic hand hardware, and help close the gap between the two. What you will do:
Research and implement reinforcement learning (RL) and other learning-based policies for robotic grasping.
Develop simulation environments to train and evaluate policies for multi-finger robotic hands.
Use tactile and joint feedback to adapt grip to different objects and disturbances, such as slip.
Benchmark learned policies against conventional control techniques.
Responsibilities
Policy Learning: Design, train and tune control policies for grasping, grip force control and coordinated finger motion.
Simulation: Build and calibrate physics-based models of robotic hands and contact to train and validate policies.
Hardware Validation: Deploy policies on robotic hand prototypes, analyze test data, and resolve differences between simulation and hardware results.
Prototyping: Work with team members on hardware setup, mechanism improvements and test fixtures.
Communication: Document methods and results, share findings with the team, and contribute to patents and publications.
Minimum Requirements
Currently pursuing PhD degree in Mechanical Engineering, Electrical Engineering, Computer Science, Robotics or related field with a background in reinforcement learning, robot learning or control
Cumulative 3.0/4.0 GPA or higher
Preferred Qualifications
Strong foundation in reinforcement learning, imitation learning or control theory.
Experience with robotics physics simulators (e.g. MuJoCo, Isaac Lab, PyBullet).
Proficiency in Python and deep learning frameworks (e.g. PyTorch).
Background in mechanical design (e.g. robotic mechanisms, CAD, prototyping) is a strong plus.
Hands-on experience with robot hardware, sensors or real-time experiments.
Creative thinker with hands-on problem solving and data-driven experimentation skills.
Works well with others and can brainstorm potential solutions together.
We are a global semiconductor company that designs, manufactures and sells analog and embedded processing chips for markets such as industrial, automotive, personal electronics, enterprise systems and communications equipment. At our core, we have a passion to create a better world by making electronics more affordable through semiconductors. This passion is alive today as each generation of innovation builds upon the last to make our technology more reliable, more affordable and lower power, making it possible for semiconductors to go into electronics everywhere. Haviv Ilan is our Chairman, CEO and President. Learn more at TI.com.
Posted 4 hours ago. Apply today while the pile is still small.