Toy Projects

Integrated System

	F1tenth Autonomous Racing Car 2024 This project is based on F1tenth, an open-source platform for small-scale autonomous driving, prepared for the 15th F1tenth Autonomous Grand Prix on IEEE ICRA 2024.
	Quadrotor’s Planning and Control 2024 This project realized the state estimation, planning, trajectory optimization, and control of a quadrotor from scratch. [Code]
	Pick and Place Challenge 2023 First place. The project is aim at pick the static and dynamic blocks and stack them. [Recording]

Reinforcement learning

PPO in Continuous Control 2024 PPO and related AC algorithms for continuous control tasks. [Code] Tried tricks to improve the performance, showing the implementation matters and a few of them are valid in the walker environment. They are Advantage, State, Reward Normalization (positive); Reward Scaling (negative); Policy Entropy (negative); Learning Rate Decay (positive); Gradient clip (positive); Orthogonal Initialization (neutral); Adam Optimizer Epsilon Parameter (neutral); Tanh Activation Function (positive).

Perception

	Mono ORB Visual Odometry 2023 A minimum python version mono visual odometry.
	Stereo Vision 2023 Reconstructe the objects from stereo camera data. Visualization: Utilized K3D and Plotly libraries for real-time visualization of 3D reconstructions, facilitating immediate analytical insights.
	NeRF 2023 A implementation of NeRF (Neural Radiance Field) by PyTorch for rendering novel views of objects.
	Augmented Reality 2023 Rendered the virtual objects of the drill and the bottle in a video by first estimating the camera pose using the info on the AprilTag, using either P3P (3-point correspondence) or PnP (n-point correspondence) approach from scratch (nonOpenCV implementation).

SLAM

	Humanoid Robot SLAM 2023 Implement particle filter SLAM in an indoor environment using information from an IMU and a LiDAR sensor. The data is collected from a humanoid robot named THOR built at Penn and UCLA. [Video about the robot]
	Differential Wheeled Robot Lidar SLAM in Indoor Environment 2022 • Deployed PRM and RRT* indoors generating global path in Gazebo simulation environment. • Adapted MPC controller in dynamic state lattice following the trajectory, visualized by ROS Rviz. • Implement EKF against particle filter as localization.

Planning

2D Path Planning Interface on OpenStreetMap 2022 • Implemented DFS, BFS, Greedy Best First compared with A* planning algorithm on building dense 2D map. • Improved weighted A* setting cost function based on real-time traffic and road conditions. • Developed algorithm visualization GUI interface interact with mouse clicking by JAVA.

Control and Hardware

	National Engineering Practice 2021 National Silver Prize amongst over two hundred universities. After implementing an array of sophisticated control strategies for tracking and precise positioning, we discovered that a finely-tuned hard-coded speed control, especially after refining the curve trajectories, offered the most reliable performance. This experience underscored a valuable lesson: complexity isn't inherently superior; adaptability and tailoring solutions to meet specific demands are crucial. [Video] [Website]
	Grant Theft Autonomous 2023 First place. A funny mechatronics competition. In this competition, operater controlled the robot without seeing the field. [GRASP Lab Youtube Channel]
	6-DoF Stewart Platform 2024 Because of the achievement in Grant Theft Autonomous 2023, I was invited to make the 6-DoF Stewart platform for UPenn Design School.