Abstract: A method for preparing a machine learning system for translating natural language traffic rules into formal logic representations for autonomous vehicles. The system includes a pre-trained language model trained on volumes of text data to assign probabilities to sequences of words. In one preparation approach, the model is used for prompting: input for the model are natural language instructions and language traffic rules, and inputs and corresponding formal logic target outputs are presented as examples to the system so that the system will learn to generate a formal logic traffic rule when prompted with natural language instructions and a natural language traffic rule. In another preparation approach, the system is fine-tuned using training data including language traffic rules as training input and corresponding formal logic traffic rules as target output for the model. The language traffic rules include natural language rules and/or synthetic language traffic rules.

Abstract: The present invention relates to a process for preparing resin impregnated wire windings by trickling an impregnating resin onto a heated rotating winding or by dipping the heated rotating winding in a bath filled with an impregnating resin. The impregnating resin comprises an epoxy resin and an initiator for the polymerization of the epoxy resin. The initiator is either a selected compound that is activated by ultraviolet irradiation, a heat-activatible initiator, or a mixture thereof. The heat-activatible initiator is a mixture comprising at least one quarternary ammonium salt and at least one thermal radical former.