Human Robot Interactions Using Sensor Fusion Approach for Intelligent Service Robotics System

• Main Authors: Nai-Wen Chang, 張乃文
• Other Authors: Ren-C Luo
• Format: Others
• Language: en_US
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/44087075635576032524

碩士 === 國立中正大學 === 電機工程所 === 97 === Mobile robots in recent years have been gradually integrated into our daily lives. Human Robot Interaction (HRI) becomes one of important topics. Traditionally, HRI systems such as voice command and other Human-Computer Interface interact with user passively. In this research, we try to make mobile robot interact with users more in active mode. Active interaction, automatic detection and following users are essential and fundamental functions. Robot with such functions can offer better service to users. For an intelligent service mobile robot the ability to track and follow target person is desirable. A robust method for tracking and following a target person with a small size mobile robot by integrating single vision sensor and laser range finder is proposed. The laser range finder and vision sensor have their respective drawbacks. To compensate the drawbacks of each sensor we present the complementary data fusion approach – Covariance Intersection, wherein it will complement the uncertainty of each sensor measurement and enhance the reliability of human’s locating position information. The Virtual Spring Model is the control rule of mobile robot that can smoothly tracking target person. Experimental results validate the robust performance of the method. The experimental results shows the vision tracking system can detect human’s upper half-body accurately and in LRF tracking system can detect human’s legs. By above systems, we can estimate the target person’s position. The complementary data fusion approach validate that it can reduce the uncertainty of target person’ position and using the virtual spring model to control robot which proves that the robot can follow target person smoothly.