Method for learning to drive growing autonomous vehicles in real time based on driver interaction

A method for learning to drive growing autonomous vehicles in real time based on driver interaction performed by a system for learning to drive growing autonomous vehicles in real time based on driver interaction, the method comprising:

• (a) an autonomous vehicle performing autonomous driving learning based on driving guidance information generated using information received from a human driving expert via an input interface device, the human driving expert being a human operator who provides guidance to the autonomous vehicle through voice instructions and manual driving intervention, wherein the autonomous driving learning comprises converting utterance information of the human driving expert into text, classifying the converted text, matching the classified text with a list of predefined driving instruction data, and generating the driving guidance information based on the matching;
• (b) the autonomous vehicle performing external driving environment recognition matching on a result of recognition of a recognition system and a result of recognition of the human driving expert, wherein the result of recognition of the human driving expert is a record stored for a point in time at which the human driving expert looks at an external driving environment, and the external driving environment recognition matching is performed by temporally matching the result of recognition of the recognition system with the record stored for the point in time;
• and
• (c) the autonomous vehicle performing reviewing and learning after driving training to improve a level of autonomous driving, wherein information on a situation in which control of driving is taken through intervention of the human driving expert is recorded from a point at which an intention of the human driving expert to take control begins to a point at which control is returned to the autonomous vehicle, and the reviewing and learning are performed using the recorded information, information of determination on whether there is an abnormality from the human driving expert, and a result of reconfirmation of commands by the human driving expert.