Design Of A Hierarchical Intelligence Architecture Based On Multiple Intelligences

Zhao Zhang,Shili Sheng,Rensong Liu,Peipei Song,Feng Duan,Hajime Asama
DOI: https://doi.org/10.1109/ICARM.2016.7606966
2016-01-01
Abstract:In this paper, we proposed a hierarchical multiple intelligences architecture consisting of three intelligences: biology intelligence, space intelligence and robot intelligence. This new intelligence architecture aims to address defects of existing intelligence architecture for controlling mobile robots. The performers of the three intelligences are human, intelligent space and robot, respectively. Human sends commands according to his or her intentions and requirements. The intelligent space integrates and provides comprehensive spatial information to assist the robot in the space. The robot performs tasks with the help of intelligent space to meet requirements of people. The proposed intelligence architecture was verified in a simulated experiment platform in which a robot accomplished two tasks including bringing a bottle and bringing an apple. A contrast experiment was conducted using the control mode that the robot was controlled by the user combining additional robot intelligence in the same simulated experiment platform. Experimental results show that the performance of the hierarchical intelligence mode is better than the contrastive control mode without space intelligence.
What problem does this paper attempt to address?