A foot-wearable interface for locomotion mode recognition based on discrete contact force distribution. (December 2015)

• Record Type: Journal Article
• Authors: Chen, Baojun; Wang, Xuegang; Huang, Yan; Wei, Kunlin; Wang, Qining
• Is Part Of: Mechatronics. Volume 32(2015)
• Page Start: 12
• Page End: 21
• Publication Date: 2015-12
• Subjects: Wearable sensing -- Contact force -- Locomotion mode recognition -- Human–machine interface -- Robotic assistive device; Computer integrated manufacturing systems -- Periodicals; Flexible manufacturing systems -- Periodicals; Mechatronics -- Periodicals; Productique -- Périodiques; Fabrication, Systèmes flexibles de -- Périodiques; Mécatronique -- Périodiques; Computer integrated manufacturing systems; Flexible manufacturing systems; Mechatronics; Periodicals; 629.89
• DOI: 10.1016/j.mechatronics.2015.09.002
• Languages: English
• ISSNs: 0957-4158

Abstract

Abstract: The knowledge of plantar pressure distribution is important in understanding human locomotion activities, and its integration with robotic assistive devices is an important potential application. In this paper, we aim to explore the potential of using discrete contact force distribution signals for locomotion mode recognition. A foot-wearable interface comprising a pair of sensing insoles, each with four sensors at selected locations, has been designed to record discrete contact forces during locomotion. Based on the information of discrete contact force distribution, we present a locomotion mode recognition strategy with decision tree and linear discriminant analysis classifiers. To verify the measurement performance of the sensing system, experiments are carried out to investigate the system stability in long term working conditions and its adaptation to different ground surfaces. To evaluate whether discrete contact force signals can be used for locomotion mode recognition, five able-bodied subjects and one amputee subject are recruited and asked to perform six types of locomotion tasks. With the proposed recognition strategy, reliable recognition performance is obtained. The average classification accuracy is 98.8% ± 0.5% for able-bodied subjects and 98.4% for the amputee subject, which is comparable to those obtained from systems based on other sensors. These experimental results indicate that monitoring of discrete contact force distribution is valuable for locomotion mode recognition. Its use can be combined with other sensing systems to achieve better performance of locomotion mode recognition for intelligent assistive device control.