mapping method for circumstances of robot using a nerve network and evolutionary computation

摘要

The present invention relates to a method for the mobile robot to perform the creation and location-aware map of the environment at the same time (Simultaneous Localization and Mapbuilding, relates to a SLAM), and more particularly to an environment map generating method of a robot using a neural network and genetic algorithms to create the map, and location awareness of the environment by using a neural network and genetic algorithms at the same time. ; of a Gaussian white noise model based, so to create a map by adjusting the connection weights of the neural network in accordance with the best noise model of the robot and the sensor according to the environment map generating method of a robot using a neural network and genetic algorithms in accordance with the present invention The noise even though there is an advantage that you can create an accurate map. In addition, to determine whether the predicted characteristic points and the actual observed feature points like the feature points with each other, even if there is an effect that the procedure to obtain the correct map.