Xuechao Zhang
Hi there! I'm currently a PhD student at IRIS Lab, ASU. With a background in both hardware and software, my passion lies in enhancing robots’ perceptiveness and dexterity, aiming for a future where they seamlessly extend human capabilities. I enjoy getting my hands dirty, finding great joy in designing and building customized robots from scratch.
When not coding or tinkering with robots, I enjoy running, swimming, and cycling (sounds like a triathlon training😼).
Github
/
G. Scholar
/
LinkedIn
/
E-mail
Affordance-Driven Next-Best-View Planning for Robotic Grasping
Xuechao Zhang, Dong Wang, Sun Han, Weichuang Li, Bin Zhao, Zhigang Wang, Xiaoming Duan, Chongrong Fang, Xuelong Li, Jianping He
Conference on Robot Learning (CoRL) 2023
Webpage
•
Paper
•
Video
2023.06
We conducted research on robotics grasping problem in cluttered desktop environments. We utilizes a parallel-gripper and incorporates NeRF (Neural Radiance Fields) for enhanced pick-and-place performance across various object categories.
2022.09
We designed a cluster of Mecanum wheeled robots for transporting blocks/slopes and building multi-layer structures.
2022.03
We developed a multi-robot testbed that exploits the ideas of digital-twin system.
2020.12
I recreated an open source four-legged robot project and submitted my contribution. This project includes motion control of a 3D printed spot micro robot, including sit, stand, angle and walk control.
2020.11
I designed a self-balancing bicycle robot, measuring 450×120×240mm in size, which was manufactured using 3D printing and utilizes an IMU to obtain its own posture, while a flywheel system is used for balance.
2020.11
We designed a desktop object-grasping robot that uses DexNet for grasp pose prediction algorithms. We trained our own neural network and explored the use of RGBD sensors and the ROS framework.
2020.09
We designed a Gomoku (Five-in-a-Row) robot that utilizes a suction cup as the end effector for a robot arm to accomplish the pick-and-place of game pieces. The system employs a camera and computer vision algorithms for piece recognition and utilizes the AlphaZero_Gomoku strategy for decision-making.
2020.04
We constructed a small forklift with a Mecanum wheel chassis in the Webots simulation environment. We implemented functions such as the vehicle’s automatic navigation, task planning, and the pick-and-place of goods.
2019.10
We designed an intelligent car with differential drive, measuring 340×245×165mm in size, which is powered by supercapacitors, capable of wireless charging at 30W through self-made circuits, and utilized inductors to detect alternating currents for navigation.
2019.08
We designed a Mecanum wheeled chess-playing robot, measuring 470×480×960mm in size, which utilizes IMU and cameras for self-positioning, an electromagnetic system for moving chess pieces, and algorithms for solving the Eight Queens problem and playing the Quoridor.
2019.03
We designed a two-wheeled differential drive robot capable of obtaining its own posture using an IMU and achieving upright balance through a PID control algorithm. It utilizes a camera to recognize its surroundings and accomplish automatic navigation.
2015.07
An international robot competition using LEGO Mindstorms technology.
∞