Method and processor circuit for consumption optimization of fully automated or partially automated driving maneuvers of a motor vehicle, motor vehicle equipped accordingly, and system

A method for consumption optimization of fully automated or partially automated driving maneuvers of a motor vehicle that are executed by at least one driver assistance system in a fully automated or partially automated manner, the method being performed based on a self-learning system operated by a processor circuit, the method comprising:

• receiving driving situation data that describe a driving situation in which a driving maneuver is executed;
• providing the driving situation data to a model of machine learning as input variables;
• generating, based on the model, a specification signal for the at least one driver assistance system from the input variables, wherein the at least one driver assistance system controls at least one actuator that controls motorized steering, an engine, or a braking system of the motor vehicle, and wherein the specification signal specifies a specification to be adhered to during the driving maneuver;
• receiving consumption data of a consumption measurement that indicate an energy requirement and/or performance requirement and/or fuel requirement incurred, and/or an emission of at least one load when executing the driving maneuver;
• mapping the consumption data on an evaluation signal by way of an evaluation function;
• signaling a training algorithm that trains the model by way of the evaluation signal, if, for the driving situation, the specification is to be retained or enhanced or avoided;
• receiving implementation data indicating that the specification has been rejected by the at least one driver assistance system;
• and
• signaling a rejected specification to be avoided by way of the evaluation signal.