RESEARCH
A Spherical Robot with Rolling and Leaping Capability
Undergraduate researcher at Bio-inspired Robotics Lab (BioRoLa), NTU (Adviser: Prof. Pei-Chun Lin)
To improve robots’ mobility on both flat ground and rough terrain, our group designed and implemented an innovative spherical robot with combined rolling and leaping capability.
-
Applied Lagrangian mechanics to analyze the robot’s system dynamics to estimate desire jumping angle
-
Involved in mechanical design, simulation of combined motion, mechatronics, control, and system integration
-
Developed an innovative energy-storing mechanism to achieve state-of the art 2 times radius(2R) jumping capability
Publications:
-
W. J. Chang, C. L. Chang, J. H. Ho, and P. C. Lin*, “Design and implementation of a novel spherical robot with rolling and leaping capability,” submitted to Mechanism and Machine Theory (MMT)
-
W. J. Chang, C. L. Chang, J. H. Ho, C. C. Tseng, and P. C. Lin*, “Design and implementation of a novel spherical robot with rolling and leaping capability,” in Proc. International Conference on Advanced Robotics and Intelligent Systems (ARIS), Aug. 2020, Taipei, Taiwan
Awards:
-
Golden Award, Allring 2020 Innovation Creative Competition
-
First place, 2020 Chinese Institute of Engineers Outstanding Student Paper Contest
Automatic Ping-Pong Eye-in-hand Robotic Arm
Final Project in Introduction to Robotics course, scored A+
Since the camera was installed right on the paddle, the challenge is to detect the ping-pong's motion in real-time less than 0.3 second. The strategy was inspired from human to first move downward the paddle and then prepare to hit the ball. The final result was to continuously hit the ping pong automatically for 12 times.
• Used multithreading and image moment to improve fast speed real-time control from 10fps to 30fps
• Responsible for Inverse Kinematics and implementing Kalman filter to estimate the system’s state
A Propeller-driven Autonomous Car
NTUME Capstone Project Formula Air
Collaborate with 4 students to achieve obstacle avoidance and adaptive cruise control
Developed a sensor fusion system that integrates multiple sensors to ensure the robustness of the vehicle.
Designed a PID controller adapted to fast response situations and won 1st place in the final competition
-
Ranked 1st place in the final race competition
AI Cup 2020 - AIWEN Mango Grade Classification Machine Learning Competition
Our group applied several Computer Vision (CV) techniques. We implemented Mask RCNN for image segmentation and Contrast Limited Adaptive Histogram Equalization (CLAHE) for highlighting defects. For the training model, I combined combined Convolutional Neural Networks (CNN) with Support Vector Machine (SVM) to classify the mangos.
-
This competition aims to automate the quality-grading process and reduce the labor costs of local farmers
-
In the final stage, the uniform blending techniques boosted our accuracy to 82%.
Novel Scale-like NiTi Blood Clot Retriever for Acute Stroke
Undergraduate researcher in Advanced Medical Device Laboratory (Adviser: Prof. Hao-Ming Hsiao)
Stroke is the second leading cause of death in the world. The main purpose of this project is to deal with acute ischemic stroke, which occurs when the blood supply to the brain is blocked by blood clot. Mechanical retriever is an effective way to treat acute ischemic stroke patients immediately compared to drug method
-
We Innovated an irregular surface shape to take out thrombus effectively with 82.8 % retrieving efficiency (Better than current retrievers 60-70%)
-
Design(SolidWorks), Simulation(Abaqus), Manufacturing(Laser cut, heat treatment)
Awards:
-
Third place, 2018 National Taiwan University System Mechanical Project Competition
-
Honorable mention, 2018 Material Innovation Award
