Optimization Strategy of Cultural IP and CAD Collaborative Design Based on Multi-Intelligent System

The multi-intelligent system can be applied to the creative generation of cultural IPs, helping designers better understand the target audience, grasp market trends, and design more attractive cultural products. Meanwhile, by integrating with CAD collaborative design systems, rapid implementation and iteration of design ideas can be achieved, enhancing the practicality and market competitiveness of the design. The design of cultural IP can help enterprises improve their competitiveness in the market and increase the trust and emotion between them and consumers. At the same time, the design of cultural IP has to meet the needs of enterprises and the emotional needs of consumers, so it needs to be communicated by multiple participants to achieve the final design effect. Previous design approaches are mostly centred on the designer, who communicates with different participants, which is time-consuming and inefficient. Therefore, this paper constructs a cultural IP and collaborative design (Co-design) optimization model based on a multi-intelligent system through a deep reinforcement learning algorithm and multi-agent cultural algorithm. The experimental results show that the model combines deep reinforcement learning, attention mechanism, and weighted computation to effectively improve the information fusion and graph information interaction performance among the intelligences, ensuring the efficiency and accuracy of participants' information interaction. In addition, the multi-agent cultural algorithm realizes the CAD collaborative purpose, improves the model convergence speed, shortens the waiting time of the participants, enhances the design efficiency, and the majority of the participants recognize the cultural IP design. © 2024 U-turn Press LLC,.