This study proposes the concepts of “behavior simple” and “behavior composed.” Behavior simple means primitive behavior, and behavior composed is a combination of behaviors simple. An artificial creature first learns some behaviors simple. Then, it learns behavior composed as a combination of behaviors simple, responding to the change of environment. This concept is applied to an Artificial Flying Creature (AFC). The AFC learns two types of flight independently: flapping and gliding. After that, it learns a sophisticated behavior by alternatively selecting these behaviors. Simulation results prove that adequate flight occurs by alternatively using flapping and gliding.
This paper proposes an artificial flying creature by flapping in a virtual air environment obeying physics law. For this purpose, a concise air drag computation method is introduced. The air drag plays the role of the environment force against the creature. The motion of the creature is automatically computed by use of the physics modeling software system and visualized as an animation movie. Results show that the creature with several pair of wings can fly by flapping as "a life as it could be".
This study aims at & n acquisition of the stable 且 ying under the virtual air environment that models the drag and the buoyancy . The desigILed modehs inspired by ex 重 stiIkg creatules with severa 【wings , and controlled by the artificial neural networkSimulations by these mode 正 s are carried out to evolve the proper behaviors by genetic algorithm , Simulation results verify that these models can 且 y stably by using wingstroke and g } iding under the Virtual air envirenment .
This study aims at acquirements of an object ( animated rob t ) motion under the air envh ・ onIIIeIlt based on physics modeling ." / e introduce scveral resista . nce forces of the air intQ physics modehng as the air environmeIlt , Silnulatiolls n fa 】 ling objects alld gliding one iIl the air verify that t} /eir Illovement reflects with the realistic world phenomenon in the a
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.