Changjuan Jing

Citation

Thesis Title
Real Time Visual SLAM for Mobile Robot Navigation and Object Detection

Read article at Massey Research Online:

In order to understand and interact with the world, a robot needs to interpret the environment in which it navigates. Essential tasks for the autonomy of a mobile robot include accurate pose estimation of the robot and map building of the environment. Ms Jing researched visual Simultaneous Localisation and Mapping (SLAM) for mobile robot navigation and object detection. She used visual SLAM techniques and terminology for robot self-localisation, conducted an evaluation and comparison of widely used RGB-D cameras, and developed a strategy for maximising robot self-localisation accuracy in a classic Extended Kalman Filter framework, a method of combining visual SLAM and object detection. Her research provided an innovative and practical approach for efficient and accurate robot navigation and will be used for balance time spent on environment observation and robot navigation, and can be further developed to enable an autonomous robot against human participants in competition environments.

Supervisors
Professor Johan Potgieter
Professor Ruili Wang
Dr Frazer Noble