• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2023, Vol. 59 ›› Issue (11): 84-93.doi: 10.3901/JME.2023.11.084

• 特邀专栏:智能产品交互设计 • 上一篇    下一篇

扫码分享

支持形态自定义和自感知的可变形界面设计方法

周子洪1, 尤伟涛1, 陆胤瑜1, 郗宇凡1, 王可幸1, 王冠云1, 孙凌云1,2   

  1. 1. 浙江大学-南方电网人工智能创新联合研究中心 杭州 310027;
    2. 浙江大学计算机辅助设计与图形学国家重点实验室 杭州 310027
  • 收稿日期:2021-12-01 修回日期:2022-06-22 出版日期:2023-06-05 发布日期:2023-07-19
  • 通讯作者: 尤伟涛(通信作者),男,1989年出生,博士,特聘研究员,主要研究方向为智能设计和审美计算。E-mail:weitao_you@zju.edu.cn
  • 作者简介:周子洪,男,1997年出生,博士研究生。主要研究方向为人机交互和智能产品设计。E-mail:zihongzhou@zju.edu.cn;陆胤瑜,男,2000年出生,硕士研究生。主要研究方向为智能产品设计。E-mail:buliugu6@outlook.com;郗宇凡,女,2000年出生,硕士研究生。主要研究方向为可持续设计。E-mail:xiyufan_echo@163.com;王可幸,女,1998年出生,硕士研究生。主要研究方向为工业设计。E-mail:wakexi328@163.com;王冠云,男,1987年出生,博士,研究员,博士研究生导师。主要研究方向为数字化设计与制造、实体交互界面、4D打印等。E-mail:guanyun@zju.edu.cn;孙凌云,男,1981年出生,博士,教授,博士研究生导师。主要研究方向为人工智能、设计智能、信息与交互设计。E-mail:sunly@zju.edu.cn
  • 基金资助:
    科技创新2030-“新一代人工智能”重大资助项目(2018AAA0100703)

Shape-changing Interface Design with the User-defined Form and Self-sensing Capability

ZHOU Zihong1, YOU Weitao1, LU Yinyu1, XI Yufan1, WANG Kexing1, WANG Guanyun1, SUN Lingyun1,2   

  1. 1. Zhejiang University-China Southern Power Grid Joint Research Centre on AI, Hangzhou 310027;
    2. State Key Laboratory of CAD & CG at Zhejiang University, Hangzhou 310027
  • Received:2021-12-01 Revised:2022-06-22 Online:2023-06-05 Published:2023-07-19

摘要: 可变形界面是一种新型的人机交互界面,智能时代下如何应用新兴技术推动可变形界面朝着智能化方向发展已成为人机交互界面研究的前沿。通过分析可变形界面的三大构成元素及其相互关系,明确了感知载体与交互方式是促进智能技术融入可变形界面的重点。围绕界面形态的自定义设计,探索了将模块化惯性传感器网络融入可变形界面的设计过程;围绕界面形态的自感知设计,探索了利用机器学习的方法实现界面形变动作的精准识别,并设计了相应的快速原型平台以提高设计效率。为评估自感知设计方法的有效性,就不同形态界面的形变动作识别结果做了对比分析;为评估形态自定义设计方法的可用性,就传感器网络的实时性、传感器硬件能耗以及硬件成本做了实验和讨论。最后,给出了可变形界面在柔性可穿戴设备、智能交互玩具、新型输入控制设备三个领域中的设计实例,表明上述设计方法能够有效开发具有相应形态功能的可变形界面,并推动可变形界面的应用场景朝更加多元化的方向发展。

关键词: 可变形界面, 传感器网络, 机器学习, 快速原型

Abstract: Shape-changing interface is a new-type human-computer interface. In the era of intelligence, how to apply emerging technologies to promote the development of shape-changing interfaces towards intelligence has become the research frontier of the human-computer interface. By analyzing the three components of the shape-changing interface and their interrelationships, it is clear that the perceptron and the interaction are the key points to promote the integration between the intelligent technologies and the shape-changing interface. For the shape-changing interface design with the user-defined form, the process of integrating the inertial sensor network into the shape-changing interface based on the modular design method is investigated. For the shape-changing interface design with the self-sensing capability, a general process of accurate deformation recognition based on machine learning is introduced and a rapid prototyping platform is developed to improve the design efficiency. To evaluate the effectiveness of shape-changing interface design with the self-sensing capability, a comparative analysis is made on the results of deformation recognition in different interfaces. To evaluate the availability of the shape-changing interface design with the user-defined form, three experiments are conducted on the real-time performance of the sensor network, the energy consumption of the sensor hardware, and the hardware cost. Finally, three design examples of the shape-changing interface in the fields of flexible wearable devices, smart interactive toys, and new-type input control devices are introduced, which show that the above design methods can effectively develop the shape-changing interfaces with corresponding shape and function, and also broaden the shape-changing application areas towards more diversified scenarios.

Key words: shape-changing interface, sensor network, machine learning, rapid prototyping

中图分类号: