Journal
Scientific and technical journal of information technologies, mechanics and optics
UDK004.4'242
Issue:2 (72)
Download PDF0 Kbyte
A genetic programming method for induction of finite state machines with continuous and discrete output actions is suggested in the article. Instead of the known approach using modeling as a controlling automaton quality estimation and demanding much time we use method of automata behavior comparison with human controlling behavior. Special feature of this approach is the opportunity to use controlling objects not only with discrete but with continuous parameters. This approach is illustrated by the example of finite state machine creation for unmanned airplane controlling in the “loop” mode.
AND ARTIFICIAL INTELLIGENCE TECHNOLOGIES GENERATION OF FINITE STATE MACHINES FOR UNMANNED AIRPLANE CONTROLLING