用于助老伴行机器人的老年人摔倒预测方法研究
|
摘要
为了使助老伴行机器人更好地服务老年人户外行走的需要,提出了一种用于助老伴行机器人的基于BP神经网络的多传感器信息特征融合进行老年人摔倒预测的方法。首先,通过老年人摔倒机理分析,提出了一种用于助老伴行机器人的老年人摔倒预测总体设计方案。然后,分别采用触觉力传感器、躯干三轴加速度计和和陀螺仪采集使用者的手部触觉力信息、躯干三轴加速度和角度信息。其次,对所采集的三类摔倒信息进行相应的特征提取,将3种特征信息采用BP神经网络进行信息融合,获取摔倒发生的概率,且当摔倒概率超过设定的阈值即判定老年人将要摔倒。最后,通过实验系统搭建和实验验证,结果表明,摔倒预测方法可靠,其整体识别准确率为97.5%,其中摔倒样本识别准确率95%,正常样本识别准确率100%,所以,该方法可以对老年人使用助老伴行机器人完成户外行走提供保证。
A prediction method of elderly fall-down was proposed based on BP network and multi-sensor information feature fusion,which could help elderly-assistant and walking-assistant robot offer better service for elderly outdoors. Firstly,based on the analysis of elderly fall-down mechanism,an overall prediction scheme for elderly is proposed. Then,touch sensors,torso three-axis accelerometers,and gyroscopes are used to collect the user's hand touch information,torso tri-axial accelerations,and angle information,respectively. The corresponding features were extracted from the collected fall-down information in three types,and fused by BP neural network to obtain the fall-down probability. When the probability exceeds the threshold,the elderly is about to fall. Finally,through experimental system construction and experimental verification,the results show that the method of fall prediction presented in this paper is reliable and the overall recognition accuracy is 97.5%,in which the fall recognition accuracy is 95% and the normal recognition accuracy is 100%. So the method can help the elderly to use the assistant robot to complete walking.
A prediction method of elderly fall-down was proposed based on BP network and multi-sensor information feature fusion,which could help elderly-assistant and walking-assistant robot offer better service for elderly outdoors. Firstly,based on the analysis of elderly fall-down mechanism,an overall prediction scheme for elderly is proposed. Then,touch sensors,torso three-axis accelerometers,and gyroscopes are used to collect the user's hand touch information,torso tri-axial accelerations,and angle information,respectively. The corresponding features were extracted from the collected fall-down information in three types,and fused by BP neural network to obtain the fall-down probability. When the probability exceeds the threshold,the elderly is about to fall. Finally,through experimental system construction and experimental verification,the results show that the method of fall prediction presented in this paper is reliable and the overall recognition accuracy is 97.5%,in which the fall recognition accuracy is 95% and the normal recognition accuracy is 100%. So the method can help the elderly to use the assistant robot to complete walking.
引文
[1]DUBOWSKY S,GENOT F,GODDING S,et al.PAMM-a robotic aid to the elderly for mobility assistance and monitoring:A“helping-hand”for the elderly[C].IEEE International Conference on Robotics and Automation,2000:570-576.
[2]WASSON G,GUNDERSON J,GRAVES S.An assistive robotic agent for pedestrian mobility[C].International Conference on Autonomous Agents AGENTS’01,2001:344-351.
[3]平伟.智能助行机器人概念样机设计[D].上海:上海交通大学,2009.PING W.Intelligent assistant robot concept prototype design[D].Shanghai:Shanghai Jiaotong University,2009.
[4]SHI F,CAO Q,LENG C,et al.Based on force sensingcontrolled human-machine interaction system for walking assistant robot[C].IEEE Transaction on IntelligentControl and Automation.2010:6528-6533.
[5]ZHANG L,CAO Q X,LENG C T,et al.The development of walking assistant robot for the elderly[J].Key Engineering Materials,2011,467-469:1893-1898.
[6]ZHU XX,CAO Q X,TAN H B,et al.Omni-direction vision based tracking and guiding system for walking assistant robot[J].Lecture Notes in Electrical Engineering,2010,67:537-543.
[7]SHI F,CAO Q X,LENG C T,et al.Based on force sensing-controlled human-machine interaction system for walking assistant robot[C].Proceedings of the 8th IEEE World Congress on Intelligent Control and Automation,2010:6528-6533.
[8]ZHANG Z,CAO Q X,LENG C T,et al.A simplified modular petri net for the walking assistant robot[J].Lecture Notes in Electrical Engineering,2010,67:529-536.
[9]FANG F.Motion control system design of a walking aid robot[D].Zhejiang:Zhejiang University College of Electrical Engineering,2010.
[10]WEI X J,ZHANG X D,WANG Y X.Research on a detection and recognition method of tactile-slip sensation used to control the elderly-assistant&walking-assistant robot[C].Proceedings of the 8th IEEE International Conference on Automation Science and Engineering,2012:1040-1045.
[11]WEI X J,ZHANG X D,YI P.Design of control system for elderly-assistant&walking-assistant robot based on fuzzy adaptive method[C].Proceedings of ICMA,2012.
[12]ZHANG X D,WANG Y X,WEI X J.Research on control technology of elderly-assistant&walking-assistant robot based on tactile-slip sensation[J].Engineering Sciences,2013,11(1):89-96.
[13]魏晓娟.助老助行机器人的触滑觉驱动技术研究[D].西安:西安交通大学,2012.WEI X J.Research on tactile awareness drive technology of assistant robot[D].Xi'an:Xi'an Jiaotong University,2012.
[14]LAI Z F,ZHANG X D,YIN G.Research on key technology of multifunctional elderly-assistant&walkingassistant robot[C].Proceedings of the 13th International Conference on Ubiquitous Robots and Ambient Intelligence,2016:767-771.
[15]LAI Z F,ZHANG X D,YIN G,et al.Research on driving control technology of a multifunctional elderlyassistant&walking-assistant robot[C].International Conference on Mechatronics and Automation Engineering,2016:189-195.
[16]MU X Q,ZHANG X D,LAI Z F,et al.Structural characteristic analysis of multifunctional elderly-assistant and walking-assistant robot based on solidworks/simulation[C].Proceedings of the 14th International Conference on Ubiquitous Robots and Ambient Intelligence,2017:941-946.
[17]HAN H J,ZHANG X D,MU X Q.An approach for fuzzy control of elderly-assistant&walking-assistant robot[C].Proceedings of the 14th International Conference on Ubiquitous Robots and Ambient Intelligence,2017:263-267.
[18]吴天昊.基于3轴加速度传感器及陀螺仪的老年人摔倒识别[D].北京:北京工业大学,2013.WU T H.Old people fall recognition based on 3-axis accelerometer and gyroscope[D].Beijing:Beijing University of Technology 2013.
[19]佟丽娜,宋全军,葛运建.基于时序分析的人体摔倒预测方法[J].模式识别与人工智能,2012,25(2):273-279.TONG L N,SONG Q J,GE Y J.Human fall prediction method based on time series analysis[J].Pattern Recognition and Artificial Intelligence,2012,25(2):273-279.
[2]WASSON G,GUNDERSON J,GRAVES S.An assistive robotic agent for pedestrian mobility[C].International Conference on Autonomous Agents AGENTS’01,2001:344-351.
[3]平伟.智能助行机器人概念样机设计[D].上海:上海交通大学,2009.PING W.Intelligent assistant robot concept prototype design[D].Shanghai:Shanghai Jiaotong University,2009.
[4]SHI F,CAO Q,LENG C,et al.Based on force sensingcontrolled human-machine interaction system for walking assistant robot[C].IEEE Transaction on IntelligentControl and Automation.2010:6528-6533.
[5]ZHANG L,CAO Q X,LENG C T,et al.The development of walking assistant robot for the elderly[J].Key Engineering Materials,2011,467-469:1893-1898.
[6]ZHU XX,CAO Q X,TAN H B,et al.Omni-direction vision based tracking and guiding system for walking assistant robot[J].Lecture Notes in Electrical Engineering,2010,67:537-543.
[7]SHI F,CAO Q X,LENG C T,et al.Based on force sensing-controlled human-machine interaction system for walking assistant robot[C].Proceedings of the 8th IEEE World Congress on Intelligent Control and Automation,2010:6528-6533.
[8]ZHANG Z,CAO Q X,LENG C T,et al.A simplified modular petri net for the walking assistant robot[J].Lecture Notes in Electrical Engineering,2010,67:529-536.
[9]FANG F.Motion control system design of a walking aid robot[D].Zhejiang:Zhejiang University College of Electrical Engineering,2010.
[10]WEI X J,ZHANG X D,WANG Y X.Research on a detection and recognition method of tactile-slip sensation used to control the elderly-assistant&walking-assistant robot[C].Proceedings of the 8th IEEE International Conference on Automation Science and Engineering,2012:1040-1045.
[11]WEI X J,ZHANG X D,YI P.Design of control system for elderly-assistant&walking-assistant robot based on fuzzy adaptive method[C].Proceedings of ICMA,2012.
[12]ZHANG X D,WANG Y X,WEI X J.Research on control technology of elderly-assistant&walking-assistant robot based on tactile-slip sensation[J].Engineering Sciences,2013,11(1):89-96.
[13]魏晓娟.助老助行机器人的触滑觉驱动技术研究[D].西安:西安交通大学,2012.WEI X J.Research on tactile awareness drive technology of assistant robot[D].Xi'an:Xi'an Jiaotong University,2012.
[14]LAI Z F,ZHANG X D,YIN G.Research on key technology of multifunctional elderly-assistant&walkingassistant robot[C].Proceedings of the 13th International Conference on Ubiquitous Robots and Ambient Intelligence,2016:767-771.
[15]LAI Z F,ZHANG X D,YIN G,et al.Research on driving control technology of a multifunctional elderlyassistant&walking-assistant robot[C].International Conference on Mechatronics and Automation Engineering,2016:189-195.
[16]MU X Q,ZHANG X D,LAI Z F,et al.Structural characteristic analysis of multifunctional elderly-assistant and walking-assistant robot based on solidworks/simulation[C].Proceedings of the 14th International Conference on Ubiquitous Robots and Ambient Intelligence,2017:941-946.
[17]HAN H J,ZHANG X D,MU X Q.An approach for fuzzy control of elderly-assistant&walking-assistant robot[C].Proceedings of the 14th International Conference on Ubiquitous Robots and Ambient Intelligence,2017:263-267.
[18]吴天昊.基于3轴加速度传感器及陀螺仪的老年人摔倒识别[D].北京:北京工业大学,2013.WU T H.Old people fall recognition based on 3-axis accelerometer and gyroscope[D].Beijing:Beijing University of Technology 2013.
[19]佟丽娜,宋全军,葛运建.基于时序分析的人体摔倒预测方法[J].模式识别与人工智能,2012,25(2):273-279.TONG L N,SONG Q J,GE Y J.Human fall prediction method based on time series analysis[J].Pattern Recognition and Artificial Intelligence,2012,25(2):273-279.