Tianci Wen

Currently, i am focusing on world models and 3D Gaussian splatting (3DGS). I have a strong interest in 3D vision, robot vision, and embodied AI (perception).

Since July 2021, I have been pursuing a five-year Ph.D. program at Nankai University, under the supervision of Professor Yongchun Fang. From 2017 to 2021, I spent an unforgettable time completing my undergraduate studies at Jilin University.

Email  /  CV  /  Google Scholar  /  GitHub

Highlights

LIVER
SLAM in underground mine

Research
Scaffold-SLAM: Structured 3D Gaussians for Simultaneous Localization and Photorealistic Mapping
Tianci Wen, Zhiang Liu, Biao Lu, Yongchun Fang*,
under review
arxiv

Scaffold-SLAM significantly outperforms state-of-the-art methods in photorealistic mapping quality across monocular, stereo, and RGB-D datasets.

Highly grateful for the kindly support from Zhiang Liu.

LIVER: A Tightly Coupled LiDAR-Inertial-Visual State Estimator With High Robustness for Underground Environments
Tianci Wen, Yongchun Fang*, Biao Lu, Xuebo Zhang, Chaoquan Tang
RA-L with IROS , 2024, oral
video / poster

Carefully coupled multi-senors lets the robot localization and mapping robustly in real-world underground mines.

Highly grateful for the kindly support from Xuebo Zhang, Zhixing Song, and Chaoquan Tang.

Robot projects
A Tightly Coupled LiDAR-Inertial-Visual State Estimator for Snake Robot
Tianci Wen, Wei Huang, Yihan Chen, Yongchun Fang*
Start from Dec. 2023

Discription: We simultaneously integrated a LiDAR, camera, and IMU into a space of approximately 10 cubic centimeters. Various gaits and speeds of the snake robot were tested, all achieving high-precision localization and mapping (with errors within 2 cm compared to motion capture).

A Navgation Framework for wheeled-leg robot
Tianci Wen, Biao Lu, Yongchun Fang*
From the winter of 2021 to the summer of 2022

Discription: We implemented an indoor navigation framework for the wheeled-leg robot based on the ROS move_base framework and LiDAR-Inertial Odometry (LIO).

Development of an L3 Intelligent Crane Operation Control System (Xuzhou Engineering Machinery Industry 'Key Technology Research' project)
Tianci Wen, Yihan Chen, Shiwei Ren, Yongchun Fang*, Biao Lu
Since the end of 2023

We are currently working on the software and hardware development, as well as the implementation, of a SLAM system for a giant industrial crane.

Seniors

Zhixing Song: The guide in SLAM and research

Haiming Gao: The guide in research direction

Early-stage papers in control


Tianci Wen, Yongchun Fang*, Biao Lu. Neural network-based adaptive sliding mode control for underactuated dual overhead cranes suffering from matched and unmatched disturbances. Auton. Intell. Syst. 2, 1 (2022).

Tianci Wen, Yongchun Fang*, Biao Lu. Adaptive antiswing control for underactuated dual overhead crane system using neural network. Acta Automatica Sinica, 2023, 49(1): 111−121