基于灰色理论的人体步态预测

HUMAN GAIT PREDICTION BASED ON GRAY THEORY

目前用户在虚拟环境中改变所处位置的方法主要是通过鼠标和键盘。针对用户在虚拟现实环境中交互性差和沉浸感不强的问题,设计一套踏板式行走装置。为了保持行走装置和用户步态的同步性,首先,提出一种基于灰色理论的人体步态预测方法,指出传统的灰色预测在形成公式时默认初始预测数值为已知条件是不合理的,应当根据实际情况来选择要使用的数据。其次,给出基于灰色理论的更新算法,将已知数据序列中每一个数据作为初始预测值建立公式模型分别计算结果,对比结果精度选择出最优预测模型。实验结果表明,在数据有限的情况下,改进型预测公式可以很好地给出预测结果,有利于踏板式行走装置与人体步态运功的同步性提高。

Currently,most users change locations by using mouse and keyboard in virtual environment. It is essential to design a pedal walking device to improve the interaction and immersion in virtual reality. In order to synchronize the walking device and the human tread movement,a human gait prediction method is proposed based on grey theory at first.However,it is unreasonable to set the initial prediction values of the formulas as default for the traditional grey prediction method. It is essential to select the useful data according to the actual situation. Next,the improved algorithm is provided based on the grey theory. The equations are derived where the known data series are selected as the initial prediction values. The best optimal prediction model is chosen according to the prediction accuracy. The experimental result shows that the improved prediction model can obtain better prediction results. It can improve the synchronicity of the pedal walking device and the human tread movement.