Qingkai Lu
As an applied scientist at Amazon, I have been working on computer vision and robotics currently. Previously I was a research scientist at Baidu Research, where I worked on deep learning, reinforcement learning, and robotics. I obtained my PhD from School of Computing at the University of Utah, under the supervision of Prof. Hermans. My PhD research focused on deep learning and robotic manipulation, especially robotic grasping. Before joining Utah, I worked on computer vision at Oregon State University and received my Master's degree in 2015.
Email: luqingkai (at) gmail (dot) com
Expertise:
Robotics, Computer Vision, Deep Learning, and Machine Learning
Publications:
Qingkai Lu*, Yifan Zhu*, and Liangjun Zhang. Excavation Reinforcement Learning Using Geometric Representation. IEEE Robotics and Automation Letters (RA-L) with IROS presentation, 2022.
Qingkai Lu and Liangjun Zhang. Excavation Learning for Rigid Objects in Clutter. IEEE Robotics and Automation Letters (RA-L) with IROS presentation, 2021. Project page.
Qingkai Lu, Mark Van der Merwe, and Tucker Hermans. Multi-Fingered Active Grasp Learning. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020 . Project Page.
Qingkai Lu, Mark Van der Merwe, Balakumar Sundaralingam and Tucker Hermans. Multi-Fingered Grasp Planning via Inference in Deep Neural Networks. IEEE Robotics & Automation Magazine (RAM) Special Issue: Deep Learning and Machine Learning in Robotics, 2020. Project Page.
Mark Van der Merwe, Qingkai Lu, Balakumar Sundaralingam, Martin Matak and Tucker Hermans. Learning Continuous 3D Reconstructions for Geometrically Aware Grasping. IEEE International Conference on Robotics and Automation (ICRA), 2020. Project Page.
Qingkai Lu, Tucker Hermans. Modeling Grasp Type Improves Learning-based Grasp Planning. IEEE Robotics and Automation Letters (RA-L) with ICRA presentation, 2019. Project page.
Qingkai Lu, Kautilya Chenna, Balakumar Sundaralingam, and Tucker Hermans. Planning Multi-Fingered Grasps as Probabilistic Inference in a Learned Deep Network. International Symposium on Robotics Research (ISRR), 2017. Project page.
Sheng Chen, Zhongyuan Feng, Qingkai Lu, Trevor Fiez, Alan Fern, Sinisa Todorovic. Play Type Recognition in Real-world Football Video. Applications of Computer Vision (WACV), pages 652–659, 2014.
Technical Skills:
C++, Python, TensorFlow, Pytorch, ROS, Gazebo, MoveIt!, OpenCV, PCL, Linux/Unix.
Internships:
Applied Scientist Intern at Amazon Robotics, Jun 2018 - Aug 2018.
Software Development Intern at Amazon (A9), May 2016 - Aug 2016.
Source Code:
Learning-based grasp planners: learning-based multi-fingered grasp planners.
Robotic grasping pipeline: a pipeline of grasping experiments and data collection for both simulation and real robot.
MoveIt configuration package: a MoveIt configuration package for the four-fingered Allegro hand mounted on a Kuka LBR4 arm.
Robot camera calibration: a calibration package for KUKA lbr4 and kinect2/asus xtion pro camera. Calibration uses Torstein A. Myhre's awesome calibration code. My main contribution of this package is the data collection part.