M.S. in Mechanical Engineering

Massachusetts Institute of Technology

• GPA: N/A
• Relevant Coursework: Visual Navigation for Autonomous Vehicles, Computer Vision, Underactuated Robotics, Robotic Manipulation, Robotics, Computational Design & Fabrication, Algorithms, Dynamics and Controls

B.S. in Mechanical Engineering

Massachusetts Institute of Technology

• GPA: 4.9/5.0
• Concentration in Control, Instrumentation, Robotics
• Minor in Computer Science, Minor in Chinese
• NEET Autonomous Machines Certificate
• Tau Beta Pi Honor Society, Pi Tau Sigma Honor Society

Research Assistant

MIT Mechatronics Research Lab

• Constructed rig with UR5 robot arm and adjustable cameras for testing object detection and manipulation algorithms
• Evaluated and visualized effects of cropping on YOLOv7 object detection using a large-scale dataset and Python

Graduate Engineering Intern

Rotor Technologies

• Achieved compression ratios up to 50:1 on raw LiDAR data by adapting H.265 video encoding using FFmpeg, as a result of researching and comparing cutting-edge LiDAR compression schemes for low-latency streaming applications
• Streamlined testing and development of perception pipelines by implementing loading and saving of both video and packet capture files using OpenCV and Libpcap to emulate virtual sensors, eliminating the need for physical sensors
• Developed an application to evaluate the effects of video encoding parameters, such as bitrate, on raw flight footage, enabling dynamic parameter adjustments in the future, which can further reduce bandwidth consumption by up to 3x
• Enhanced flexibility and modularity of LiDAR pipeline, simplifying the integration of new sensors and processes
• Created a streaming app that features optional azimuth-level decimation and supports Velodyne and Robosense LiDARs

Undergraduate Engineering Intern

Rotor Technologies

• Spearheaded expansion of subscale helicopter testing infrastructure, used to evaluate perception, communications, data logging, and flight controls software for remotely-piloted full-scale helicopters, tripling the number of available units
• Constructed CAD model of subscale helicopters to obtain physical parameters for simulation and future SAS tuning
• Wired all power and data connections among servos, mini-PC, flight board, RC receiver, and other onboard components
• Defined and bridged EKF state and airspeed sensor data messages from Pixhawk PX4 to proprietary flight software
• Directed the purchasing of parts and hardware for subscale helicopters and coordinated with suppliers in Taiwan

Undergraduate Assistant

MIT New Engineering Education Transformation

• Soldered and assembled 30 RF controllers and 50 custom Arduino Mega sensor shields for student-built robot cars
• Guided 40+ students on using switches, infrared sensors, ultrasonic sensors, color sensors, and IMUs with Arduino

Research Intern

NCTU Computer Games and Intelligence Lab

• Developed C++ script to play the game 2048 using temporal difference reinforcement learning
• Obtained the winning 2048-tile after 2,000 training rounds and 93% win rate after 20,000 training rounds