Advances in cognitive robotics, achievements and challenge

Abstract

The contemporary robotics technology is broadening its applications from factory to more general-purpose applications in domestic and public use, e.g., partner to the elderly, rehabilitations, search and rescue, etc. If robotics technology is to be successful in such complex, unstructured, dynamic environments with high level of uncertainties, it will need to meet new levels of robustness, physical dexterity and cognitive capability. This presentation discusses an emerging field called cognitive robotics. The one solution for building cognitive robots in order to cope with imprecise, incomplete, and inconsistent information that arises in complex technical systems, is computational intelligence that uses biologically inspired soft-computing techniques, like artificial neural networks, evolutionary approaches, and swarm intelligence. Research topics, features and challenges of cognitive robotics will be introduced. Key challenges in constructing these robots include the systematic treatment of uncertainties, the modeling of the environmental state, the coordination of teams of cooperating robots in dynamic environments, the interaction with humans, development, and learning. It is important to notice that in order to realize cognitive robots many overlapping disciplines are needed, e.g. robotics, artificial intelligence, cognitive science, neuroscience, biology, psychology, and cybernetics. Some important research topics from this area will be specially analyzed: Advanced perception (vision, tactile sensing, haptic sensing, multi-sensor fusion), Advanced locomotion and manipulation, SLAM, Learning including imitation learning, reinforcement learning, supervised learning, Human-robot interaction, Reasoning and Making Decisions, Intelligent planning and navigation, Swarm intelligence, etc. A case study of cognitive methods applied for humanoid and service mobile robots will be introduced.