Qinjie Lin

I am currently a fourth-year Phd student in the Computer Science Department at Northwestern University, advised by Prof. Han Liu.

I received my Bachelor's Degree in 2018 at South China University of Technology, where I was advised by Prof. Sheng Bi.

I completed an AI Research Scientist internship at Meta Reality Lab and a Machine Learning Engineer Internship at Zebra Tech.

Email / CV / Google Scholar / Linkedin / Github

Research

My research interests lie in the general area of Robotics, particularly in LLM for robotics and augmenting LLM with planning.

	EMS®: A Massive Computational Experiment Management System towards Data-driven Robotics Qinjie Lin, Guo Ye, Han Liu, accepted on ICRA, 2023 submitted paper / website / code / video We propose EMS®, a cloud-enabled massive computational experiment management system supporting high-throughput computational robotics research.
	RoboFlow: a Data-centric Workflow Management System for Developing AI-enhanced Robots Qinjie Lin, Guo Ye, Jiayi Wang, Han Liu, CoRL, 2021 paper / website We propose RoboFlow, a cloud-based workflow management system orchestrating the pipelines of developing AI-enhanced robots.
	Switch trajectory transformer with distributional value approximation for multi-task reinforcement learning Qinjie Lin, Han Liu, Biswa Sengupta Submission to ICML, 2023 paper We propose SwitchTT, a multi-task extension to Trajectory Transformer but enhanced with two striking features: (i) exploiting a sparsely activated model to reduce computation cost in multi-task offline model learning and (ii) adopting a distributional trajectory value estimator that improves policy performance, especially in sparse reward settings.
	JBDL: A JAX-Based Body Dynamics Algorithm Library for Robotics. Cheng Zhou, Lei Han, Yuzhu Mao, Guo Ye, Qinjie Lin, Wenbo Ding, Han Liu, Zhaoran Wang, Zhengyou Zhang Working on submission github We develop a JAX-Based body dynamics algorithm library for rigid body dynamics. This library contains a highly efficient python library that contains some essential rigid body dynamics algorithms.
	Collision-free Navigation of Human-centered Robots via Markov Games Guo Ye, Qinjie Lin, Tzung-Han Juang, Han Liu, ICRA, 2020 paper / video/ code Collision-free navigation for human-centered robots is interesting and we exploit adversarial learning to learn robot collision avoidance policy.
	Learning to Plan in High Dimensions via Neural Exploration-Exploitation Trees Binghong Chen, Bo Dai, Qinjie Lin, Guo Ye, Han Liu, Le Song, ICLR, 2020 arXiv/ video/ code Path planning in high dimension is pretty tricky but we propose a meta path planning algorithm to solve planing problems.
	Tower Stacking on Baxter Robot Arm Jiarui Li, Qinjie Lin, Shufeng Ren, Patricia Sung, Yuchen Wang, Northwestern University ME 495: Embedded Systems in Robotics, fall 2018 project/ video/ code Using Baxter robot arm to pick different blocks on the table and stack them together.
	Indoor Mapping Using GMapping on Embedded System Qinjie Lin, Zhaowu Ke, Sheng Bi,Sirui Xu, Yuhong Liang, Fating Hong, Liqian Feng, ROBIO*, 2017 paper We improve performance in time consumption and CPU consumption by designing and implementing a mapping system based on Embedded platform.
	Optimization of Robot Path Planning Parameters Based on Genetic Algorithm Yuhong Liang, Fating Hong, Qinjie Lin, Liqian Feng, Sheng Bi, RCRA, 2017 paper We Propose adopting the genetic algorithm to optimize parameters for local path planning of mobile robot.
	A Global Localization System for Mobile Robot Using LIDAR Sensor Liqian Feng, Sheng Bi, Min Dong, Fating Hong, Yuhong Liang, Qinjie Lin* and Yunda Liu IEEE-CYBER, 2017 paper Addressing mobile robot localization problems for robots and creating 2D occupancy grid map from LIDAR sensor data.