End-to-end Visual Policy Learning

End-to-end Visual Policy Learning

Currently open to new students?

Yes
No

Description

The goal is to obtain a lightweight policy that can run in real time on the drone to avoid obstacles and reach a goal. We will look into SoTA VLA models, RL models, and diffusion policies, and test how they run. We will set up large-scale training data and train our own models.

Skills Desired

Required
  • Strong programming
  • Strong communication, collaboration, organization, and planning
  • Proactive and fast self-learner
Bonus
  • Familiarity with simulation software (IsaacSim, AirSim, Gazebo)
  • Familiarity with machine learning: PyTorch
  • Familiarity with software: ROS2, Python, C++, Linux, Bash, Git

Student Learning Objectives

  • Learn and strengthen the skills above
  • Understand the full pipeline required to develop an intelligent robot from start to finish
  • Gain hands on experience with drones

Classes Accepted into Project

Junior
Senior
Graduate Student
Not a hard limit. We care more about talent and potential.

Compensation

Units 9
Units 12
Pay

Contact