Deep Reinforcement Learning in Robotics Navigation
- Lei Tai, Giuseppe Paolo, Ming Liu, Virtual-to-real Deep Reinforcement Learning: Continuous Control of Mobile Robots for Mapless Navigation, (Under Review) arXiv video .
LIDAR Odometry Demo
Left shows multi-sensors mounted on a car, Velodyne VLP-16, Occam Omni Camera IMU and GPS are equipped.
Right is and a demo of large-scale LIDAR Odometry. The data is recorded in Hong Kong, from Hang Hau to HKUST. Total size of the map is larger than 3km x 0.9km x 0.1km.
Author: Haoyang Ye, Yuying Chen, Ming Liu
Deep-learning-based human-like decision-making and exploration
- Lei Tai, Shaohua Li, and Ming Liu, A Deep-network Solution Towards Model-less Obstacle Avoidence, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Daejeon, Korea, 2016 PDF ,
Deep Visual Homing throug Omnidirectional Camera
- Lei Tai and Ming Liu, A Technical Report on Deep Visual Homing through Omnidirectional Camera, 2016 PDF ,
Asynchronous Deep Reinforcement Learning
One of the first real Cloud Robotic Systems
A remote controlled mobile robot based on IntoRobot Cloud (www.intorobot.com) and IntoRobot Atom board. The control of the robot is through Internet. The robot can be shared to other via WeChat and WhatsApp easily.
Omnidirectional Camera Calibration Toolbox
This is a calibration toolbox for omnidirectional cameras which can be used with MATLAB. The most common omnidirectional camera models can be used for calibration and the toolbox has good corner extraction capabilities even for quite distorted omnidirectional images.
The toolbox is based on LIBOMNICAL.
This toolbox fixes some bugs, adds some additional features and was mainly used and tested with the Mei model. Therefore the other model implementations need probably some bug fixes too and everybody is welcome to contribute!
The idea is to have a general omnidirectional camera calibration toolbox for all the common models and a platform for future model implementations.
More Information see our GitHub
Tango-based application (on-going)
libcnn: A library for real-time robotic recognition applications based on CNN
libcnn is a modular deep learning libraray, useful for robotics and computer vision. More Information see our GitHub
Visible Light Communication-based Localization and Path-planning
Localization using a photonic sensor mounted on a tablet. It reaches to high localization precision as shown in the video. Please refer to:
Kejie Qiu, Fangyi Zhang, Ming Liu, Visible Light Communication-based Indoor Localization using Gaussian Process, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Hamburg, Germany, 2015
Ming Liu, Kejie Qiu, Shaohua Li, Fengyu Che, Liang Wu, C. Patrick Yue, Towards Indoor Localization using Visible Light Communication for Consumer Electronic Devices, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2014
Multi-robot 3D SLAM based on LSD-SLAM (collaboration with ETH - ongoing)
The individual maps generated by LSD-SLAM over multiple robots are fused.
Multi-robot 2D SLAM without known initialization
Multiple robots will move across unknown environments, so that a complete map will be constructed once the co-localization can be achieved.
Internet-of-things control of a flowerpot
Online Extrinsic calibration between extroceptive Sensors
Guoyang Xie, Tao Xu, Carsten Isert, Michael Aeberhard, Shaohua Li, Ming Liu, Online Active Calibration for a Multi-LRF System, IEEE 18th International Conference on Intelligent Transportation Systems (ITS), 2015
Visual Homing-based topological Navigation
The mobile robot navigates in an indoor environment with illuminant changes by using an Omnidirectional Camera. Please refer to:
Ming Liu, Cedric Pradalier, Roland Siegwart, Visual Homing from Scale with an Uncalibrated Omnidirectional Camera, IEEE Transactions on Robotics (TRO), Vol. 29, Issue 6, pp.1353 - 1365, Dec. 2013.
3D modeling and mapping in real-time
This a demo of 3D registration by ICP, based on the library proposed in:
Francois Pomerleau, Stephane Magnenat, Francis Colas, Ming Liu, Roland Siegwart, Tracking a Depth Camera: Parameter Exploration for Fast ICP, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2011
Localization by 1-bit sensor for educational Robots
This is a joint work with EPFL, U Leuven, INRIA and CityU. We localize an educational robots using a inferred sensor on gray-scale images.
Shiling Wang, Francis Colas, Ming Liu, Francesco Mondada, Stéphane Magnenat, Localization of inexpensive robots with low-bandwidth sensors, DARS 2016, PDF
Point-cloud analysis for semantic labelling using Tensor Voting
Ming Liu, Francois Pomerleau, Francis Colas and Roland Siegwart, Normal Estimation for Pointcloud using GPU based Sparse Tensor Voting, IEEE International Conference on Robotics and Biomimetics (ROBIO), 2012,
Ming Liu, Efficient Segmentation and Plane Modeling of point-cloud for structured environment by Normal Clustering and Tensor Voting, in Proceedings of the IEEE International Conference on Robotics and Biomimetics, (ROBIO) 2014
Ming Liu, Roland Siegwart, Information Theory based Validation for Point-cloud Segmentation aided by Tensor Voting, Best Paper in Information IEEE International Conference on Information and Automation (ICIA), 2013
Incremental regional topological segmentation for environment surveillance
Ming Liu, Luc Oth, Francis Colas, Roland Siegwart, Incremental Topological Segmentation for Semi-structured Environments, Autonomous Robots (AURO), Vol. 37, Issue 3, Aug. 2014.
Color-based description of environment and topological scene recognition
Ming Liu, Roland Siegwart, Topological mapping and scene recognition with lightweight color descriptors for omnidirectional camera, IEEE Transactions on Robotics (TRO), Vol. 30, Issue 2, pp. 310- 324, April. 2014.
Resource allocation among networked robots using real-time wireless communication
Matthew Tan, Ming Liu and Roland Siegwart, An Experimental Evaluation of the RT-WMP Routing Protocol in an Indoor Environment, Best Paper Finalist IEEE International Conference on Information and Automation (ICIA), 2013Last modified on Tue, Nov 17, 2015