基于感性工学和BP神经网络的服务机器人头部形态设计研究

目的产品的形态设计要素和用户对产品的感性意向评价之间的关系很难被准确表达。在服务机器人产品的形态设计中,将用户的感性需求准确表现到最终的方案设计中一直是亟待解决的难题。方法运用感性工学中的语义差异法和BP神经网络算法,对服务机器人头部形态的设计要素与用户的感性意向评价之间的关系进行分析。结果最终拟合出这两者之间的映射关系,并在此基础上运用VB语言将训练好的BP神经网络模型接入三维建模软件中,构架出服务机器人头部形态的辅助造型设计系统。结论使得造型过程更加理性,极大地提高了设计效率。下一步的工作应建立更科学、全面的服务机器人辅助造型设计系统,包括将本研究成果扩大到服务机器人的整体外形设计中;将不同的设计约束条件对造型产生的影响作为BP神经网络的输入层纳入到拟合过程中等。

Head Shape Design of Service Robots on Kansei Engineering and BP Neural Network

It''s difficult to accurately express the relationship between the product design elements and the user''s perceptual knowledge of the product. It''s a problem in how to display the user''s perceptual demand accurately to the final design scheme in the form design of service robots product. In this study, the relationship between the design elements of the head shape of the service robots and the perceptual intention of the users was analyzed by using the semantic difference method and the BP neural network. Finally, the mapping relationship between them was fitted, and the aided design system of service robots'' head shape was structured by using VB language in connecting the trained neural network model to 3D modeling software. It makes the design process more rationally and greatly improves the design efficiency. The next step should be to establish a more scientific and comprehensive aided design system of service robots (including expanding the results to the overall design of the service robots) in order to incorporate the influence of different design constraints on the shape as the input layer of the BP neural network into the fitting process.