Some Evaluations of Efficiency of Machine Synthesis of Terrain Images Affecting Learning Process Using Simulators for Vehicle Drivers

Abstract

Software appeared in recent years that allows almost anyone to start drawing, does not always meet the requirements for obtaining images of the terrain in the simulators. When synthesizing the visual 3D model of a part of the visualization scene, observed through the window of the simulator's cabin, it is necessary to obtain a model of space in which the observer can train his eye. The quality of this model is to allow a student to acquire professional driving skills. The authors take into account that a student acquires professional skills in driving a real vehicle by driving the model of the vehicle moving in a model of space. When learning using the appropriate simulator, possible deflections maximize the acquisition of socalled "false skills". Studies have shown that there are always false skills when learning on a simulator. This learning for driving a vehicle on a simulator differs from learning driving on a real vehicle. When learning using a real vehicle, false skills are not acquired. The analysis showed that one of the reasons for the appearance of false skills in learning driving a vehicle based on information received from a human visual system are errors in synthesizing a picture on the screen of forming intermediate picture. The article considers some types of errors related to an incorrect display of shadows when synthesizing the 3D models and suggests ways of eliminating them.