Adaptive quasi-Monte Carlo method for uncertainty evaluation in centroid measurement of planetary rovers. (February 2021)

• Record Type: Journal Article
• Authors: Na, Qiang; Hu, Shurong; Tao, Jianguo; Luo, Yang
• Is Part Of: Transactions of the Institute of Measurement and Control. Volume 43:Number 3(2021)
• Page Start: 623
• Page End: 634
• Publication Date: 2021-02
• Subjects: Uncertainty evaluation -- adaptive quasi-Monte Carlo method -- Monte Carlo method -- planetary rover -- centroid measurement; Automatic control -- Periodicals; Measuring instruments -- Periodicals; Commande automatique -- Périodiques; Mesure -- Instruments -- Périodiques; 681.2
• DOI: 10.1177/0142331220950038
• Languages: English
• ISSNs: 0142-3312

Abstract

The measurement of the centroid is of great significance to improve the control performance and reduce the energy consumption of the planetary rover (PR). The uncertainty is an essential indicator of the reliability of centroid measurement results. The purpose of the current study is to evaluate the uncertainty of centroid measurement in the multi-configuration rover. For the measurement of the centroid, the model with 37 parameters of two measurements as the input and the centroid coordinates as the output is derived. Further, the mechanical and electrical integrated system is developed, which can measure the centroid of PRs in different configurations and sizes. Moreover, to overcome the shortcomings of the Monte Carlo method (MCM) in uncertainty evaluation, an adaptive algorithm that automatically determines the number of input sequences is proposed. On this basis, an adaptive quasi-Monte Carlo method (AQMCM) is presented in order to accelerate the uncertainty evaluation, which is characterized by the randomized Sobol sequence. Besides, experiments are performed to compare the uncertainty evaluation process and results of the AQMCM and the adaptive Monte Carlo method (AMCM) in multiple configurations. The result shows that the standard uncertainty of the AQMCM is almost the same as that of the AMCM, but the sequence size of AQMCM is evidently smaller than that of AMCM. Taken together, we identify that the AQMCM evaluates the uncertainty of CM for the multi-configuration rover in an efficient and fast way. Furthermore, the AQMCM provides a new method for uncertainty evaluation, particularly for nonlinear models in different states.