Individuation and Interaction Enhancement of Landscape Design Based on Computer-Aided Design and Reinforcement Learning

The traditional approach to landscape design primarily hinges on the designer's expertise and gut feeling, greatly constraining the design's adaptability to diverse and personalized user needs. Current research, therefore, focuses on integrating cutting-edge technology into landscape design to enhance its individuality and user engagement. This article aims to pioneer innovative methods that enhance the uniqueness and interactivity of landscape designs, thus catering to the escalating demands of users. To this end, we introduce a novel approach that seamlessly merges Computer-Aided Design (CAD) with Reinforcement Learning (RL) algorithms. This approach allows us to capture users' preferences in real-time and leverages RL technology to continuously refine the design proposal, fostering a profound interaction between users and the design process. Rigorous experimental validations and user evaluations reveal that this approach significantly outperforms traditional design methodologies in terms of user satisfaction, design efficacy, and an overall superior score. The findings underscore that this method not only elevates the individuality and interactivity of designs but also propels the automation and intelligence of the design process. © 2024 U-turn Press LLC.