Open AccessAccuracy improvement of positioning data in greenhouse for agricultural machinery via optimisation algorithm
Author(s) -
Geng Jing,
Wang Xiang,
Zhu Saihua,
Lin Xiangze
Publication year - 2019
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2018.9378
Subject(s) - centroid , computer science , position (finance) , algorithm , agricultural machinery , raw data , greenhouse , agriculture , artificial intelligence , ecology , finance , horticulture , programming language , economics , biology
Although several indoor positioning systems for greenhouse have been developed in automated agricultural machinery monitoring, the data measured by positioning systems still has large errors. In order to solve the problem, in this study, improved weighted‐centroid algorithm and maximum likelihood estimation (MLE) algorithm were developed to optimise positioning data of agricultural machinery in the greenhouse, which is measured by ultra‐wide band‐based indoor positioning system according to different agricultural machinery working states. Compared with real coordinate data, when agricultural machinery stays still, the average static error of the data was 66.4 mm with the improved weighted‐centroid algorithm; when the machinery is in dynamic state, the average dynamic error of the data was 61 mm and the probability that error <60 mm was 0.848 according to the MLE. Hence, the optimisation algorithms presented in this study improve the positioning accuracy significantly and the position of agricultural machinery is able to be estimated better compared with the raw data.