摘要
为了使智能下肢能够实时地调节膝关节阻尼,使其固有频率与实际步频相符,达到更加接近正常步态的效果,提出了一种基于蚁群搜索理论的卡尔曼步频预测算法.利用蚁群算法在规定的连续域中寻找卡尔曼预测方程中系统误差与测量误差两个参数的最优组合,实现对卡尔曼预测方程的优化.通过获取六组人行走的步频数据进行实验,结果表明:经蚁群算法优化后,卡尔曼预测算法预测得到的步频与后验值之间的误差比跟随方式的误差分别减少了44.10%,43.42%和36.17%,证明了基于蚁群搜索理论的优化卡尔曼步频预测方法在智能下肢控制上具有良好的应用前景.
In order to make the intelligent lower limb be able to adjust the damping of the knee joint in real time,and make its natural frequency be in conformity with actual pace,so as to achieve the most natural gait,a optimal Kalman stride frequency prediction algorithm based on the theroy of ant colony search was proposed in this paper.The stride frequency was predicted by the Kalman prediction equation.And the optimal combination of two parameters in Kalman prediction equation,the system error and the measurement error,was found in a continuous domain by using the ant colony algorithm.6groups of the stride frequency data were obtained.The final test results show that in the 3groups of test data,the deviation between the next step stride frequency predicted by the Kalman prediction equations which is optimized by the ant colony algorithm and the posterior values reduce 44.10%,43.42% and 36.17%,comparing with the following method.It prove that the method combined the Kalman prediction algorithm and ant colony algorithm have an application prospect on controlling the intelligent lower limb.
出处
《华中科技大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2015年第7期109-113,共5页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
福建省自然科学基金资助项目(2012J01413)
中央高校基本科研业务费专项资金资助项目(XMU.2013121018)
大学生创新创业训练计划资助项目(XMU.DC2013009)
关键词
智能下肢
足底压力
步频
卡尔曼预测方程
蚁群算法
intelligent lower limb
plantar pressure
stride frequency
Kalman prediction equations
ant colony algorithm