Learning Behavioral Knowledge In Robotic Domains
Marco Botta, C. Barogio, A. Giorclana, B. Graziano · 2005
This paper presents a system which learns and maintains a body of heuristic rules, useful to decide about the behaviour of a robot, from apriori knowledge and a set of examples. The apriori knowledge consists of a causal model of the domain, stating the relation- ships among basic phenomena, and a body of phe- nomenological theory, describing the links between ab- stract concepts and their possible manifestations in the world. The phenomenological knowledge is used deduc- tively, the causal model is used abductively and the examples are used inductively. The problems of imper- fection and intractability of the theory are handled by allowing the system to make assumptions during its reasoning. In this way, robust knowledge can be learned with limited complexity and limited number of exam- ples. For the sake of illustration, an simple example where a robot is asked to learn whether an object is dangerous because hot, or not, is presented .