Projects
Here is a list of my selected projects. Click on any project to see its details.
Autonomous Aerial Robotics
Independent Course Projects of ELEC5660 (Introduction to Aerial Robotics) @ HKUST, 2025 Spring. Instructor: Prof. Shaojie SHEN
- Developed control algorithms, trajectory planning, and sensor fusion for autonomous aerial robots by utilizing rigid-body dynamics, A* path planning, PnP-based localization, visual odometry, and (augmented state) Extended Kalman Filter (EKF)-based state estimation to enable vision-based indoor navigation and real-time flight control.
Vision and Learning-Based Robotics Autonomous Navigation and Manipulation
Independent Course Projects of ELEC4260 (Intelligent Robots and Embodied AI) @ HKUST, 2025 Spring. Instructor: Prof. Ping TAN
- Implemented LiDAR-based SLAM with ICP alignment, A* path planning, Bezier-curve trajectory generation, and PID control to achieve autonomous mobile robot navigation and exploration.
- Trained and evaluated GraspCNN, GraspCNN2, and PoseCNN models for robotic arm manipulation, designing grasp synthesis strategies and validating the grasping pipeline in simulated and real-world settings.
Respiratory Monitoring System
Group Leader, Undergraduate Final Year Project (self-proposed). Supervised by Prof. Jun ZHANG
- Award: ECE Best FYP/T Award 2024-2025 2nd Runner-Up
- Designed and implemented a hybrid VAE-LSTM-MHA model for time-series forecasting and anomaly detection of respiratory diseases (sleep apnea) in noisy patient respiration data, achieving over 95% detection accuracy and improving reconstruction quality, especially on multi-modal inputs, by at least 10% over the baseline VAE-LSTM architecture.
- Constructed a low-latency wireless communication system using ESP-NOW protocols achieving sustained end-to-end latency under 50ms to ensure robust peer-to-peer connectivity and minimal protocol overhead.
Autonomous and Fast Robots
Independent Course Projects of ECE4160/5160 (Fast Robots) @ Cornell University, 2024 Spring. Instructor: Dr. Jonathan Jaramillo
- Developed and programmed an autonomous driving framework for a fast robot car, with sensing (multi-sensor fusion), localization (Bayes filter-based localization), environment mapping, control (PID), path planning and navigation on a microcontroller board (SparkFun RedBoard Artemis Nano).
