Yiqi Lyu (吕一奇)
Email: yiqilyu AT andrew DOT cmu DOT edu
I am currently a full-time Research Assistant at the Safe AI Lab, Carnegie Mellon University.
I hold a Master's degree in Mechanical Engineering from Carnegie Mellon University,
and a Bachelor's degree in Mechanical Design, Manufacturing, and Automation from Huazhong University of Science and Technology.
Research Interest
My research intersects robotics, control, machine learning, and mechanical engineering, with a goal to bridge the gap between advanced robotics research and practical applications. I believe in a holistic approach to intelligent robotics, which integrates hardware design, control strategies, and decision-making to create comprehensive, efficient robot systems.
News & Updates
2023/06 - I will join Safe AI Lab as a research assistant.
2023/05 - I graduate from master's program in Mechanical Engineering at Carnegie Mellon Univerisy.
On-going Research
Skill-based Imitation Learning for Robotic Arm Manupulation in Dynamic Engironments
Objective: The goal of this project is to create a generalizable robotic arm manipulation system capable of operating in dynamic environments involving human interactions.
Key words: Generalizable Policy, Data-driven, Imitation Learning, Energy Function, RRT, A*, Real-world Deployment
Status: Under Preparation
Generalizable Beam Walking for Legged Robots with Reinforcement Learning
Objective: developing a learning-based framework for quadruped robot beam walking.
Key words: Quadruped Robot, Agile Locomotion, Curriculum Learning, Visuomotor Control, Gaussian Splatting
Status: Under Preparation
Selected Projects
The Phylogenetic Origin of the Red Panda's False Thumb as an Adaptation to Arboreal Locomotion
Date: Jan 2023 - May 2023
A modular, bio-inspired red panda paw is designed and fabricated with which to experimentally investigate the evolutionary reason for the existence of the false thumbs of red pandas.
Key words: Bio-inpsired Robot, CAD, FEA, Mechatronics, Actuators and Control, Arduino
Dynamic Balance and Stability in Tumbler Robotics
Date: Oct 2022 - Dec 2022
The objective of this project is to ensure the stability of a tumbler robot, maintaining its upright position during various maneuvers including standing, moving, and turning. To elevate the complexity and intrigue of the project, an additional pendulum is affixed to the top of the robot. This pendulum must be kept stable even as the robot is in motion.
Key words: MPC, Minimun-jerk Trajectory, Input Shaping, MATLAB, Simulink