The main focus of this research is to develop an autonomous robot capable of self-navigation in an unknown environment. The proposed system performs autonomous navigation primarily based on the following visually perceived information:

  • Range Estimation: A novel variable single/multi baseline omnidirectional stereovision system with automatic baseline selection (processing offloaded to the GPU). It also features a Bumblebee stereo camera for dynamic obstacle avoidance.
  • Motion Estimation: A 3DoF visual odometry system combining distance travelled estimated by a ground plane optical flow tracking system, with bearing estimated by the panoramic visual compass system.
  • Place Recognition: An appearance-based place recognition system using image signatures created from Haar decomposed omnidirectional images for loop closure detection.

These modules are integrated into the mobile robot’s navigation and mapping system which balances its effort amongst loop closing and exploration, decides its next course of action, performs topological map building and path planning, and executing the selected path. For more details, please refer to the selected publications

Selected Publications

An Omnidirectional Vision System for Outdoor Mobile Robots
W. L. D. Lui and R. Jarvis
SIMPAR Workshop on Omnidirectional Robot Vision, Venice, Italy, 2008, pp. 273-284

A Pure Vision-based Approach to Topological SLAM
W. L. D. Lui and R. Jarvis
Proceedings of the IEEE International Conference on Intelligent Robots and Systems, Taipei, Taiwan, 2010, pp. 3784-3791
DOI: http://dx.doi.org/10.1109/IROS.2010.5651115

An Active Visual Loop Closure Detection and Validation System for Topological SLAM
W. L. D. Lui and R. Jarvis
Australasian Conference on Robotics and Automation, Brisbane, Australia, 2010

Eye-Full Tower: A GPU-based variable multibaseline omnidirectional stereovision system with automatic baseline selection for outdoor mobile robot navigation
W. L. D. Lui and R. Jarvis
Robotics and Autonomous Systems, vol. 58, iss. 6, pp. 747-761, 2010
DOI: http://dx.doi.org/10.1016/j.robot.2010.02.007

Autonomous Robot Navigation: Appearance based Topological SLAM
W. L. D. Lui
PhD Thesis, Monash University, Australia, 2011
Link: http://arrow.monash.edu.au/hdl/1959.1/473850

A Pure Vision-based Topological SLAM System
W. L. D. Lui and R. Jarvis
The International Journal of Robotics Research, vol. 31, iss. 4, pp. 403-428, 2012
Link: http://ijr.sagepub.com/content/31/4/403