关键词:
Conceptual learning
Mobile robots
Baby Robot
Autonomous navigation
Multiresolutional knowledge
摘要:
In this thesis the autonomous robot is presented as a "baby" — that is, a system with no a priori knowledge of the world in which it operates, but with behavior acquisition techniques that allow it to build this knowledge from the experiences of actions within a particular environment. The learning techniques are rooted in a recursive algorithm for inductive generalization of nested schemata molded from processes of early cognitive development in humans. The algorithm extracts data from the environment and using clustering and abduction, it creates schemata interpreted as control rules. This system is robust enough to deal with a constantly changing environment; such changes provoke the creation of new schemata by further generalizing from experiences. The algorithm is capable of automatically generating the hierarchy of task decomposition, job that is performed in existing systems only by organizing the knowledge of a human-expert. The system searches in a multiresolutional knowledge base which it generates; thus, the complexity is pseudo-polynomial which ensures learnability. Although the traditional approach to autonomous navigation involves off-line path planning with a known world map, in most of the real tasks the environment is not well known because of ever-changing conditions of the assignment; absence of gravity; and sophisticated, hard to predict obstacles like components of the space stations, etc. Baby Robot gathers data from its sensors and then by using a schema-discovery system it extracts concepts, forms schemata and creates a quantitative/conceptual semantic network. Experimenting with Baby-Robot is important in unknown media because the rules of required behavior do not coincide with human intuitions. This thesis demonstrates that the simulated device leans unexpected schemata from its own experience. When the Baby Robot is first dropped onto a terrain, into space, or under water into the depths of the ocean, it does not have any experiences and i