面向VR放置任务的自然手势交互时间的预测

为了在虚拟现实（virtual reality，VR）环境中准确模拟用户的放置任务和避免由交互设备抖动等因素导致的放置精度低的问题，利用自然手势进行交互，并提出一种基于改进Fitts定律的面向VR放置任务的自然手势交互时间的预测方法。首先，构建了“手势交互—VR环境—VR交互任务”的交互逻辑，建立了自然手势与交互任务之间的语义映射关系；其次，根据自然手势设计原则，进行了面向VR放置任务的自然手势交互实验；然后，结合VR放置任务的特点，将移动距离、目标物容差和移动物体大小融入经典Fitts定律，建立了基于改进Fitts定律的自然手势交互时间预测模型；最后，对实验数据进行回归分析。回归分析结果显示：基于经典Fitts定律拟合得到的相关系数R²=0.515，交互时间的预测值与真实值存在较大偏差，表明经典Fitts定律不适用于面向VR放置任务的自然手势交互时间的预测；基于改进Fitts定律拟合得到的相关系数R²=0.970，交互时间的预测值与真实值相吻合，表明改进Fitts定律具有良好的适用性，可准确预测面向VR放置任务的自然手势交互时间。研究结果丰富了VR交互方式，并可为VR环境中人机交互界面的设计提供指导。

Prediction of natural gesture interaction time for VR placement task

In order to accurately simulate the placement task of users in VR (virtual reality) environment and avoid the low placement accuracy caused by the shaking of interaction devices, the interaction was carried out by using natural gestures. A prediction method of natural gesture interaction time for VR placement task based on improved Fitts Law was proposed. Firstly, the interaction logic of "gesture interaction-VR environment-VR interaction task" was constructed to establish the semantic mapping relationship between natural gesture and interaction task; secondly, according to the design principle of natural gesture, the natural gesture interaction experiment for VR placement task was carried out; then, combined with the characteristics of VR placement task, the moving distance, target tolerance and the size of the moving object were integrated into the classical Fitts Law, and the prediction model of natural gesture interaction time based on improved Fitts Law was established; finally, the regression analysis of the experimental data was carried out. It was found that the correlation coefficient R²=0.515 by fitting based on the classical Fitts Law, and there was a large deviation between the predicted value and the real value of interaction time, which indicated that the classical Fitts Law was not suitable to predict the natural gesture interaction time for VR placement task; the correlation coefficient R²=0.970 based on improved Fitts Law, and the predicted value of interaction time was consistent with the real value, which showed that improved Fitts Law had good applicability and could accurately predict the natural gesture interaction time for VR placement task. The research results enrich the VR interaction mode, and can provide guidance for the design of human-computer interaction interface in VR environment.