Robots

Autonomous Vehicle - A Multiple-Spectrum Sensing Platform

We are creating novel compact multiple-spectrum visual sensors to improve the autonomous driving capability indispensable of time and seasons. The novelties of these hybrid sensors include a wider-FOV angle, synchronization and pixel-wise spatial alignment of cross-spectrum channels. With these sensors, we create a huge multiple-spectrum dataset of images/videos, which were collected in both rural and urban areas, across different time of a whole day and different season of a whole year. Join us if interested in it!

Campus Guide: A 16-channel Lidar based Mobile Robot  

In this work, we DIY a mobile robot platform equipped with a 16-line Lidar and a stereo camera (not used in this work). Basically, we do everything in sensing that is necessary for navigation with a 16-line Lidar only, including 3D mapping, robot localization, traversable region detection, obstacle detection and avoidance, path and motion planning, etc. Our prototype robot can freely navigate in campus environment at an average speed of 15KM/h.  

Reference paper: 

• Minghao Liu, Zhixing Hou, Zezhou Sun, Ning Yin, Hang Yang, Ying Wang, Zhiqiang Chu and Hui Kong, Campus Guide: A Lidar-based Mobile Robot, European Conference on Mobile Robotics (ECMR) 2019 (oral presentation)