Vision and 3D map-based accurate real-time localisation system for autonomous cars



Autonomous car (Eurecar, KAIST) as a platform for this research

Autonomous cars, or self-driving cars, have been demonstrated by several companies such as Google, utilising high-end LIDAR sensor and inertial navigation system. This international collaboration project, funded by South Korean government (KAIST, University of Michigan and ANU) is focusing on enhancing absolute and relative navigation accuracy by investigating the robust perception and sensor integration, aiming for driving in complex road conditions. Specific goals of these projects are to achieve

  • 3D mapping of complex roads
  • All-source SLAM navigation
  • All-terrain navigation 


KAIST, South Korea


J. Cheng, J. Kim, W. Zhanga, J. Shaoc, Z. Jianga, “Robust Linear Pose Graph-based SLAM,” Robotics and Autonomous System, Springer (doi:10.1016/j.robot.2015.04.010) (available on-line May 2015)

Cheng, J., Kim, J., Jiang, Z., Zhang W., Tightly Coupled SLAM/GNSS for Land Vehicle Navigation, Lecture Notes in Electrical Engineering, Volume 305, 2014, pp 721-733. Print ISBN 978-3-642-54739-3

C. Jiantong, J. Kim, ACRA, Delayed Optimisation for Pose-Graph SLAM, Australasian Conference on Robotics and Automation, Melbourne, Dec. 3-5, 2014.

S. Huh, D. H. Shim, and J. Kim, Integrated Navigation System using Camera and Gimbaled Laser Scanner for Indoor and Outdoor Autonomous Flight of UAVs, IEEE International Conference on Intelligent Robots and Systems (IROS 2013), Japan, Nov. 3-8, 2013

Updated:  8 September 2015/Responsible Officer:  Dean, CECS/Page Contact:  CECS Marketing