Forecasting a gait cycle parameter region to enable optimal FES triggering. (2020)

• Record Type: Journal Article
• Authors: Parthasarathy, Aniruddha; V.N., Megharjun; Talasila, Viswanath
• Is Part Of: IFAC-PapersOnLine. Volume 53(2020)Supplement
• Page Start: 232
• Page End: 239
• Publication Date: 2020
• Subjects: Gait cycles -- Forecasts -- BJ-THR -- HSR -- CL -- updates -- triplets; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2020.06.040
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: Efficient rehabilitation for muscles/nerves using Functional Electrical Stimulation needs control energy to be delivered at exactly a desired point (gait parameter) in a gait cycle. Since both the stimulator and body muscles have a built in lag (to respond), it is important that the gait parameter be predicted ahead of time. Current work focuses mostly on computing gait parameters and triggering the electrical stimulator. In this work, we have used a second-order Box-Jenkins polynomial model in combination with a simple thresholding algorithm to forecast a region around one of the gait parameters - the Heel Strike Region (HSR) - three time steps ahead. Each set of the three forecasts were assigned a certain confidence level to indicate which of the three forecasts were in the HSR. From the results obtained, we can conclude that the HSR can be forecasted 30 milliseconds ahead of time with good accuracy.