Autonomous Drone project
This project is based on DJI M100 Drone. The project is using camera and DJI guidance integrated sensor system to navigate the drone itself and detect the toys in the box. The SDK is on ROS. And we have to create our package to do the navigation, detection and capture in full autonomy.
Atlas07 Mining Robot
The NASA Robotic Mining Competition takes place every year at the Kennedy Space Center in Florida. The goal of this competition is to design robots that are capable of collecting Martian soil and deposit it into a bin.
The Robot for NYU is called Atlas, which dig the soil by using the wheels. The robot could dig the wheel by driving the front and back wheel in opposite direction.
As a team member in the Electrical and Computer subteam, my teammate Conner and I build the ROS simulation for our Robot Atlas 07.
In the simulation, Atlas 07 could do SLAM, Path planning, obstacle avoidance using Velodyne Lidar and other sensors based on the Gmapping or AMCL algorithm.
For the real Robot, we installed the ROS system on the Raspberry Pi 3 model B and use the 2d-lidar(because the price) to implement the SLAM algorithm. Due to the broken of the shaft, the robot got stuck in the arena in 2017's competition.
Next year, we are building two robots working together autonomously and trying to use the April tag and camera to make a localization system to win the competition.
Voice Based Control Manipulator
In this project, users could speak the name and size of the shape that he or she wants to draw. And the manipulator could do the task since we write the code of the inverse kinematics of the end-effector based on the defined local frames.
This is project collaborates with Kota Tomaru.
Project | 01
Project | 02
Project | 06
Data Acquisition System for AUV
Completed a data acquisition system on Robosub for receiving and processing the signal under
water for the senior design project, which is equipped with the robot the International Robosub
Competition (2016) in San Diego, CA.
We design our own PCB hardware part to filter the analog signal in the water first then use ADC convert it into the digital signal. Finally, we use FPGA save and processing the digital signals from different hydrophones and calculate the location of the pinger.
Project | 03
Project | 04
Project | 05
Gazebo Simulation of 6 Turtlebots
The goal of this project is to simulate the control of 6 turtlebots in Gazebo that need to navigate an environment. The robots will maintain a formation using a decentralized controller. The robots have a function to know their neighbors. And only one robot has its own GPS value. All the robots could send information and receive from others.
Autonomous Ground Robot for Fruit Counting at Orchard using SLAM and Machine Learning
The aim of this project is to develop a prototype of a full autonomous ground robot that
can achieve three main goals: Simultaneous Localization and Mapping(SLAM), Navigation and
Counting the number of fruits using machine learning to estimate the production of orchard to
The robot with these functions could improve farmers’ efficiency and save