Introduction

I am currently a Phd Candidate at School of Electrical & Electronics Engineering, Nanyang Technological University (NTU), Singapore, under supervision of Professor Xie Lihua. My reseasrch interests inlcude simultaneous localization and mapping (SLAM), visual place recogntion (VPR), and path planning. My project experience includes robotics oprating system (ROS), unmanned aerial vehicle (UAV), and automated guided vehicle (AGV). Feel free to email me at wh200720041@gmail.com if you are interested in my topic, or you are looking for robotics expert.

Selected Publication (First Author)

Online Visual Place Recognition via Saliency Re-identification

July 2020
Han Wang, Chen Wang and Lihua Xie

IEEE/RSJ International Conference on Intelligent Robots and Systems 2020 (IROS)

Intensity Scan Context:

Coding Intensity and Geometry Relations for Loop Closure Detection

May 2020
Han Wang, Chen Wang and Lihua Xie

IEEE International Conference on Robotics and Automation 2020 (ICRA)

Fast Loop Closure Detection via Binary Content

June 2019
Han Wang, Juncheng Li, Maopeng Ran, Lihua Xie

IEEE International Conference on Control and Automation 2019 (ICCA)

Feasible Computationally Efficient

Path Planning for UAV Collision Avoidance

June 2018
Han Wang, Muqing Cao, Hao Jiang, Lihua Xie

IEEE International Conference on Control and Automation 2018 (ICCA)

Open Source Projects

Fast Lidar Odometry And Mapping (F-LOAM)

  • Ranked 7th in KITTI dataset
  • 20Hz processing speed

Intensity Scan Context based Lidar SLAM (ISC-LOAM)

  • Ranked 9th in KITTI dataset
  • Full SLAM structure with backend optimization

Simulation Toolkit for Warehouse Robot

  • Provide simulation of warehouse environment
  • A simple demonstration of robot localization and navigation

Loop Closure Detection via Saliency Re-identification

  • Re-formulate visual place recognition as a saliency re-identification problem
  • Increased computational efficiency and performance compared to traditional bag of word model

3D Localization and Scene Reconstruction for Solid-State Lidar

  • Localization with the latest solid-state Lidar
  • Colored scene reconstruction

Project Experience

2018 - Now
In Collabration with Delta Electronics, Taiwan
This project aims to develop a highly integrated AGV (automated guided vehicle) cooperative system for material handling and conveying in smart and reconfigurable manufacturing to meet the demand of LVHM production. The system is autonomous and intelligent in terms of optimal transportation path planning, reconfigurability, fault tolerance, and cost effectiveness.

Demo List

Autonomous AGV navigation and precision docking

Autonomous AGV trasporation with robot arm integration

Active navigation and mapping with MPC based trajectory optimization

2016 - 2018
In Collabration with ST Engineering, Singapore
The project aims to develop light-weight obstacle detection and avoidance technology for UAVs that are operating in a GPSless, indoor environment. Traditionally, obstacle detection is achieved using a LiDAR or a camera. However, LiDAR operates only in the 2D plane, has limited viewing angles and is often costly to implement. Hence, in this project, it is proposed to develop a light-weight 3D obstacle avoidance technology based on time of flight sensors such as those from Intersil. The obstacle detection and avoidance system will be integrated with the UWB based localization and navigation system to achieve safe navigation in complex GPSless, indoor environments.

Demo List

UAV collision avoidance demo under different obstacle environment

UAV collision avoidance test by Muqing Cao

2014-2015
In Collabration with STMicroelectronics, Singapore
This project aims to develop speech recognition system to manipulate drones. Most drones requires manual or remote manipulation to take off and land. Existing speech recognition requires either online training or huge computational cost. In this project an offline speech recognition algorithm is designed and implemented to STM32 on board processor. The UAV is able to listen to voice command and execute task without manual control. We won the 1st place in Singapore Amazing Flying Machine Competition.
2013 - 2014
Undergraduate Research on Campus (URECA)
The project aims to develop an intelligent task planning algorithm for container terminal operation. Epuck robot is used to simulated individual transportation truck in the container terminal map designed based on real container terminal in Singapore. The algorithm optimizes the time cost under various parallel tasks input to the system. In the meantime, collision avoidance is garanteed.

Demo List

Multi robot task distribution and collabration in simulated container terminal map

Competition & Awards

RobotX International Challange 2018

Judge's Special Award

RoboMaster 2018

2nd place in Group Stage

RobotX International Challange 2016

Team Leader, 6th Place

RobotX International Challange 2014

Best Teamwork Award

Singapore Flying Machine Competition 2016

2nd Place

Singapore Flying Machine Competition 2015

1st Place

President Research Scholar Award 2014

1st Place

SELECTED Research Resutls

Participated research projects or valuable research works selected from our research group

Receive the Best Paper Award in robotic planning at ICAR 2017
Robust Optical Flow Using Kernel Cross-Correlators, presented at ICRA 2018
GPS-less Localization and Path planning for UAV Inspection of 3D Structures
Integrated UWB-Vision Approach for Autonomous Docking of UAVs in GPS-Denied Environments
Non-intrusive WiFi-based indoor positioning systemwith a user-friendly web interface
SG50-Marina Bay Sands UAV Formation Show