计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (12): 110-116.DOI: 10.3778/j.issn.1002-8331.1811-0315

• 模式识别与人工智能 • 上一篇    下一篇

基于SAE和LSTM的下肢外骨骼步态预测方法

陈超强,蒋  磊,王  恒   

  1. 中国矿业大学(北京) 机电与信息工程学院,北京 100083
  • 出版日期:2019-06-15 发布日期:2019-06-13

Gait Prediction Method of Lower Extremity Exoskeleton Based on SAE and LSTM Neural Network

CHEN Chaoqiang, JIANG Lei, WANG Heng   

  1. School of Mechanical and Information Engineering, China University of Mining and Technology(Beijing), Beijing 100083, China
  • Online:2019-06-15 Published:2019-06-13

摘要: 提出一种基于栈式自动编码器(Stacked Auto Encoder,SAE)和长短时记忆(Long Short-Term Memory,LSTM)神经网络相结合的步态预测方法解决下肢外骨骼机器人跟随控制问题。人体在行走过程中下肢步态具有一定的周期性,通过将下肢运动信息作为输入,步态作为输出,构建SAE-LSTM神经网络模型,并利用Keras对SAE-LSTM神经网络进行搭建和验证。实验结果表明,SAE-LSTM神经网络根据之前时间段的步态序列有效地预测出下一时刻的步态信息,平均准确率能够达到92.9%以上。

关键词: 外骨骼, 步态预测, 栈式自动编码器, LSTM神经网络

Abstract: A gait prediction method based on SAE and LSTM neural network is proposed to solve the problem of follow-up control of lower extremity exoskeleton robot. Since the human body has a certain periodicity of the lower limbs posture during walking, the SAE-LSTM neural network model is constructed, by using the lower extremity motion information as inputs, the gait as an output. And using the Keras to build and validate the SAE-LSTM neural network. The experimental results show that the SAE-LSTM neural network can effectively predict the gait information at the next moment according to the previous gait sequence, and the average accuracy can reach more than 92.9%.

Key words: exoskeleton, gait prediction, stacked autoencoder, Long Short-Term Memory(LSTM) neural network