Multisensor data fusion via Gaussian process models for dimensional and geometric verification

Colosimo, Bianca Maria; Pacella, Massimo; Senin, Nicola

doi:10.1016/j.precisioneng.2014.11.011

Cited by 61 publications

(49 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the uncertainty propagation of weighted least-squares fusion has not been well investigated, although it is known that equation (6) is a BLUE [13]. It can be shown from equation (8) that, the variances of the predictions (principal diagonal elements of the covariance matrix), or the standard uncertainties, vary at different observing positions. An arithmetical mean of the variance values can usually be taken as a typical indicator representing general changes of uncertainty or accuracy level.…”

Section: Uncertainty Analysismentioning

confidence: 99%

“…The former four groups employed the proposed weighted fusion methods; while the last group employed the GPRA fusion technique highlighted in previous research work [8]. Within each group of tests, the mean value of the reconstruction errors 1 , 2 , and ∆ are presented, and the standard deviation of ∆ is also given to show the stability of the fusion performance.…”

Section: F-theta Lensmentioning

confidence: 99%

“…The GPRA method has been shown to be a successful approach to fuse the 9635 SL sample points with randomly selected 100 CMM sample points [8]. In consideration of computing time and comparability with the former research in [8], the proposed weighted fusion methods were tested here by using simple randomly selected 100 SL points and 100 CMM points as the reconstruction input.…”

Section: A Freeform Artefactmentioning

confidence: 99%

“…In consideration of computing time and comparability with the former research in [8], the proposed weighted fusion methods were tested here by using simple randomly selected 100 SL points and 100 CMM points as the reconstruction input. With 200 tests for each fusion method, the statistics of the RMS individual fitting errors ( 1 for SL data and 2 for CMM data), fusion error and the reduction of the fusion error ∆ were shown as in Figure 8.…”

Section: A Freeform Artefactmentioning

confidence: 99%

“…Most fusion methods rely on advanced surface fitting techniques under a residual approximation (RA) framework [5]. For example, in the research work [8,9], Gaussian process (GP) RA fusion (GPRA), which approximates the residuals between two inputs using a GP model, has been demonstrated to be a successful solution to the fusion of heterogeneous data. Other advanced fitting models, such as B-spline and wavelet models [10,11], may also be able to provide satisfactory fusion results based on the RA fusion framework.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Study of weighted fusion methods for the measurement of surface geometry

Wang

Pagani

Leach

et al. 2017

Precision Engineering

View full text Add to dashboard Cite

A note on versions:The version presented here may differ from the published version or from the version of record. If you wish to cite this item you are advised to consult the publisher's version. Please see the repository url above for details on accessing the published version and note that access may require a subscription. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Highlights:  Four weighted fusion methods for surface measurement were classified and analysed.  The uncertainty propagation and the relationship with Kalman filter were analysed.  Advanced models compatible with different fusion methods were described.  Experiments verified the effectiveness of weighted fusion on accuracy improvement.  Weighted fusion complements residual approximation fusion in many fusion scenarios. AbstractFour types of weighted fusion methods, including pixel-level, least-squares, parametrical and non-parametrical, have been classified and theoretically analysed in this study. In particular, the uncertainty propagation of the weighted least-squares fusion was analysed and its relation to the Kalman filter was studied. In cooperation with different fitting models, these four weighted fusion methods can be applied to a range of measurement challenges.The experimental results of this study show that the four weighted fusion methods compose a computationally efficient and reliable system for multi-sensor measurement problems, especially for freeform surface measurement. A comparison of weighted fusion with residual approximation-based fusion has also been conducted by providing the input datasets with different noise levels and sample sizes. The results demonstrated that weighted fusion and residual approximation-based fusion are complementary approaches applicable to most fusion scenarios.

show abstract

Section: Uncertainty Analysismentioning

confidence: 99%

Section: F-theta Lensmentioning

confidence: 99%

Section: A Freeform Artefactmentioning

confidence: 99%

Section: A Freeform Artefactmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Study of weighted fusion methods for the measurement of surface geometry

Wang

Pagani

Leach

et al. 2017

Precision Engineering

View full text Add to dashboard Cite

show abstract

Gaussian process regression‐based detection and correction of disturbances in surface topography measurements

Maculotti

Genta

Quagliotti

et al. 2021

Quality & Reliability Eng

View full text Add to dashboard Cite

Modern smart and intelligent manufacturing is characterised by an increasing use of highly engineered surfaces and quasi‐free form geometries, for example, by additive manufacturing, and the requirement for fast and informative measurement tools for quality controls in production. These have lately pulled towards the adoption of optical surface topography measuring instruments to qualify technological surfaces, which are core to being assessed to characterise the product and optimise the manufacturing process. Surface topography measurements performed by optical instruments are known to suffer from disturbances, such as non‐measured points and spikes. These are due to complex interactions between the measurand and the instrument and may result in poor measurement quality and biased characterisation results. Currently, the correction of measurement disturbances is carried out by empirical approaches which mostly involve thresholding and interpolations, whose sensitivity is often devolved to the operator. In this work, a Gaussian process regression‐based approach is outlined to identify and correct measurement disturbances exploiting spatial correlation properties of the measurements to provide a formal, robust, and univocally defined approach to manage measurement disturbances. The proposed approach, differently from currently available alternatives, is capable of managing at once the different type of disturbances by means of a supervised machine learning technique. The formal and practical advantages of the proposed method are discussed exploiting case studies of industrial interest applied on quasi‐flat surfaces to stress the disturbances effect.

show abstract

Sequential sampling for functional estimation via Sieve

Benevento,

Ahadi,

Gupta

et al. 2024

Quality & Reliability Eng

View full text Add to dashboard Cite

Sequential sampling methods are often used to estimate functions describing models subjected to time‐intensive simulations or expensive experiments. These methods provide guidelines for point selection in the domain to capture maximum information about the function. However, in most sequential sampling methods, determining a new point is a time‐consuming process. In this paper, we propose a new method, named Sieve, to sequentially select points of an initially unknown function based on the definition of proper intervals. In contrast with existing methods, Sieve does not involve function estimation at each iteration. Therefore, it presents a greater computational efficiency for achieving a given accuracy in estimation. Sieve brings in tools from computational geometry to subdivide regions of the domain efficiently. Further, we validate our proposed method through numerical simulations and two case studies on the calibration of internal combustion engines and the optimal exploration of an unknown environment by a mobile robot.

show abstract

Multisensor data fusion via Gaussian process models for dimensional and geometric verification

Cited by 61 publications

References 21 publications

Study of weighted fusion methods for the measurement of surface geometry

Study of weighted fusion methods for the measurement of surface geometry

Gaussian process regression‐based detection and correction of disturbances in surface topography measurements

Sequential sampling for functional estimation via Sieve

Contact Info

Product

Resources

About