Autonomous Learning in Mobile Cognitive Machines
Abstract
Intelligence is a capability ascribed typically to animals, but not usually to plants. Animals canmove while plants do not. Is the mobility a necessary condition or driving force for the emergenceof intelligence? We hypothesize that mobility plays a foundational role in evolving animal andhuman intelligence, thus, is fundamentally important in understanding and creating embodiedcognitive systems (Clark, 1996). In this project we aim to develop a new class of machine learningalgorithms for mobile cognitive systems that actively collect data by sensing and interacting withthe environment. We envision a new paradigm of autonomous AI that overcomes the previous AIparadigms of top-down/rule-driven symbolic and bottom-up/data-driven statistical systems. Inspiredby the dual process theory of mind (Kahneman, 2011), we develop a dual-memory learningarchitecture for mobile cognitive machines that combines the hippocampus and cortex models ofmemory to learn rapidly, flexibly, and robustly. We use mobile robot platforms to investigate theautonomous learning algorithms and demonstrate their capability in real-world home environments.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Sep 21, 2018
- Source ID
- FA23861614089
Entities
People
- Byoung-tak Zhang
Organizations
- Air Force Office of Scientific Research
- Seoul National University
- United States Air Force