Tutorial III intelligence and cognition for robotics

Abstract

The discussion on cognition, intelligence, consciousness, and self has a long history. Robotics can be considered as a study on understanding human behaviors and building artificial physical agents. In this tutorial, I talk about the research direction on robotics from historical point of view. First, I explain traditional problems on map building and path planning of mobile robots. One of the most important problems is simultaneous localization and mapping. Next, I explain various types of robotic researches, e.g., behavior-based robotics, cognitive robotics, and distributed robotics. The study on intelligent robotics tries to realize individual function or method related with search, reasoning, planning, recognition, knowledge acquisition, and learning, while the study on cognitive robotics deals with holistic mechanism related with cognitive phenomena such as perception, attention, reasoning, planning, memory, and learning based on constructivism. I explain three research directions of cognitive robotics; (1) cognitive development, (2) behavior acquisition, and (3) social communication. Computational intelligence including fuzzy, neural, and evolutionary computation plays the important role to realize cognitive development of robot partners from the methodological point of view. Nowadays, the synthesis of information technology, network technology, and robot technology may bring the brand-new emerging intelligence to robots from the technical point of view. We have to deal with huge size of data and information received and measured from different types of sensors, robots, smart devices, Internet, and others simultaneously. Therefore, the environment surrounding people and robots should have a structured platform for gathering, storing, transforming, and providing information. Such an environment is called informationally structured space. The informationally structured space is constructed through interdependent and complementary information sharing among robot partners and wireless sensor nodes in a house. Robot partners should have human-like intelligence and cognitive capabilities to co-exist, communicate, and interact with people. We show several examples of cognitive robotics and feasibility studies using robot partners. We have been developing different types of robot partners called iPhonoid and iPadrone using smart devices. These robot partners are interconnected through the informationally structured space each other. Finally, we discuss the future direction of intelligent robotics and cognitive robotics based on informationally structured space