This study presents a novel method for regulating the target price of a new automotive product, combining deep learning, game theory, and control theory to enhance pricing strategies. Unlike existing pricing methods, our framework uniquely combines CNN-based cost feature extraction, GAN-driven design optimization, and RL-based feedback control in a unified hyperstructure. Our approach employs Convolutional Neural Networks (CNNs) to analyze cost features extracted from Product Lifecycle Management (PLM) data, demonstrating a significant reduction in analysis time by 30% compared to traditional methods. A Generative Adversarial Network (GAN) aids in the effective management of design options, optimizing design costs by up to 25%. Reinforcement learning within a learning-based control framework enables convergence to Nash equilibrium. This integration results in optimal pricing strategies that align target costs with market demands, increasing projected profitability by 15% over standard pricing models. This research highlights the innovative application of multidisciplinary techniques in automotive pricing.