Influences of sensing positions on principal components and performance of a one-dimensional distributive tactile sensor

Songklanakarin Journal of Science and Technology (SJST)

关键词: neural network;    principal component analysis;    tactile sensing;    position determination;   

DOI  :  

来源: DOAJ

【 摘 要 】

This paper describes an arrangement of a one-dimensional distributive tactile sensing system that can be used todetermine an applied position of a point load of a constant magnitude. The performance of the system was examined usinginputs derived from a mathematical model and a back propagation neural network as an interpretation algorithm. Performancesof the system with 2–8 inputs with and without an application of principal component analysis (PCA) as a preprocessor wereexamined. For each number of inputs, four sets of sensing positions were explored and the accuracies in determining anapplied load position were compared. It was found that the system was able to determine an applied load position with errorsin the range of 1.0–3.1 mm depending on the number of inputs and the method of inputting data. The error decreased with anincrease in the number of inputs. It was found that input preprocessing by PCA impaired the performance. Systematicallychosen and optimal sets of sensing positions resulted in the most desirable performance and their performances were comparable.Amongst the sets of input positions explored, random positions yielded the highest errors. Random positions alsoresulted in the largest difference between the first two principal components.

【 授权许可】

Unknown