Jingxian Wang

Single Motor Drone Swarm System

Fall 2021 – Present

Major Collaborator: Prof. Michael Rubenstein

This research demonstrates the first single motor drone swarm capable of autonomous flight using peer-to-peer (P2P) communication and sensing. Each drone, weighing only 20g and containing just one moving part, can fly independently with full control, featuring onboard P2P communication and precise positioning enabled by a novel infrared-based system. The drone is 10x lighter than all previous systems with comparable P2P positioning accuracy.

Paper | Source Code | Video @ Bilibili

Reconfigurable Legged Metamachine

Spring 2024 – Present

Major Collaborator: Chen Yu, David Matthews, Prof. Sam Kriegman

This research introduces autonomous modular legs: agile yet minimal, single-degree-of-freedom jointed links that can learn complex dynamic behaviors and may be freely attached to form legged metamachines at the meter scale. This research enables rapid repair, redesign, and recombination of highly dynamic modular agents that move quickly and acrobatically through unstructured environments.

Paper | Webpage | Source Code

Single Motor Spherical Robot

Fall 2023 – Present

Major Collaborator: Prof. Michael Rubenstein, Chen Yu

This research presents Rollbot, the first spherical robot that maneuvers on a 2D plane using only a single motor. By modulating motor speed and an attached mass, it controls trajectory curvature to follow circular paths and waypoints. Rollbot has simpler construction and lower cost comparing to conventional multi-motor spherical robots, and can also serve as a fail-safe for multi-motor spherical robots.

Paper | Source Code | Video @ Bilibili

PCB-based Minimalist Robot PCBot

Fall 2021 – Spring 2022

Major Collaborator: Prof. Michael Rubenstein

Best paper finalist, best student paper finalist, best mechanisms and design paper finalist at IROS 2022

This work presents PCBot, a minimalist robot driven by a PCB-integrated bi-stable solenoid actuator. This actuator simplifies manufacturing and reduces assembly time to under 20 seconds. A prototype demonstrates precise motion on an orbital shake table. This research showcases the potential of minimalist, externally-driven designs in robotics.

Paper

Film Grain Rendering and Parameter Estimation

Fall 2022 – Spring 2023

Major Collaborator: Kaixuan Zhang, Prof. Thrasyvoulos N. Pappas

This research proposes a fast film grain rendering algorithm based on analytic statistics from a physics-based Boolean model used by Newson et al. Our method estimates model parameters from scanned images and enables real-time simulation when individual grains are not visible. Experiments show a six-order speed-up over Newson et al.’s Monte Carlo approach with strong visual fidelity. This work has potential applications in film production, computer graphics, and video compression.

Paper

Underwater Localization System Based on Electrostatic Field

Fall 2020 – Fall 2022

Major Collaborator: Junzheng Zheng, Xin Guo, Chayutpon Huntrakul, Prof. Chen Wang, Prof. Guangming Xie

This research presents a novel electric sense-based localization scheme for small free-swimming underwater robots in large-scale environments. Inspired by electric sensing in fish, the system uses distributed emitters and an onboard receiver to estimate position and orientation. Experiments demonstrate robust, accurate localization where vision and sonar struggle, offering a lightweight solution for constrained underwater robots.

Paper

Realistic Black Hole Accretion Disk Rendering

Fall 2019 – Spring 2020

Major Collaborator: Prof. Bin Chen

This paper introduces a double ray-tracing method for fast, physically accurate rendering of accretion disks near spherical spacetimes and used it to render the Schwarzschild black holes and JMN-1 naked singularities.

Report | Source Code | Video @ Bilibili | Video @ YouTube

Light-actuated Soft Flying Robot

Fall 2018 – Spring 2021

Major Collaborator: Dr. Shudeng Ma, Prof. Guangming Xie

This research explores light-actuated soft flying robots using liquid crystal polymer films driven by projected light. Although the method failed due to material softening under high light intensity, it revealed key insights into light-based actuation.

Construction of Optical Tweezer System

Fall 2018 – Spring 2020

Major Collaborator: Qiuyuan Wang, Prof. Jing Yang

This work focuses on the construction of an optical tweezer with a trapping strength measurement system. We also investigated the possibility for phase-adjustable SPPs optical tweezer array.

Distributed Localization Without Direct Communication Inspired by Statistical Mechanics

Fall 2016 – Fall 2018

Major Collaborator: Tianye Wang

This research proposes a distributed localization algorithm where robots estimate positions by exchanging virtual particles via sensor data only. It enables accurate, anchor-free localization and shape formation, validated in both simulation and real-world swarms.

Paper