Search

Global Navigation Satellite Systems (GNSS) easily allow precise determination of position, velocity and time (PVT) enabling hundreds of different applications in almost any field of modern life, from transportation to logistics, from surveying to disaster management, from natural resources monitoring to services for citizens. In this field, the collaboration involving the NavSAS group at Politecnico di Torino and Fondazione LINKS (former ISMB) together with the NAVIS group at HUST is lasting since more than 10 years. Besides focusing on basic research aspects that provide knowledge and skills, which are essential for achieving excellence and expertise, a considerable effort has been devoted to the development of technological solutions that allow precise and reliable PVT determination, which is crucial in land management and environmental applications. In particular, a low-cost prototype GNSS receiver named NavisA has been developed. The equipment can ensure an accuracy in the range of some centimeters and is therefore suited for land management applications. The equipment was tested in different conditions by independent investigators that verified its performances confirming that it can contribute to the modernization of the land management sector in Vietnam, where traditional time-consuming total station measurements are still widely used.

This paper deals with the experimental evaluation of contact pressure at the interface between an elastomeric rod seal for pneumatic cylinders and its metallic counterpart without interposing any intrusive measuring device. A new test bench, which is able to measure the radial force exerted by a rod seal displaced at constant velocity on a sensorized portion of a cylinder rod over time, was designed and manufactured. The seal was pressurised to reproduce actual working conditions. A data postprocessing methodology was developed for an indirect evaluation of contact pressure starting from the experimental data set of the radial force exerted by the seal on the rod. At first, the measured radial force signal was filtered and properly fitted obtaining a differentiable function; then, contact pressure distribution was computed as a function of radial force time derivative, seal velocity and rod diameter. Preliminary experimental results are presented.

Scientific collaboration between Italian and Vietnamese institutions in the field of Global Satellite Navigation Systems (GNSS) dates back to 2004, when a first train-the-trainer action was conducted in the Hanoi University of Science and Technology (HUST) by researchers of Politecnico di Torino and Istituto Superiore Mario Boella (ISMB). In the following years, a closer and tighter collaboration between these institutions has allowed for the setup of a Collaboration Centre on GNSS in HUST, the NAVIS Centre. After more than a decade of collaboration, the NAVIS Centre has become a key reference point in the area for European and Italian institutions and companies active in the field of GNSS. The paper describes the activities and projects that have allowed the set up and growth of the Centre as well as the achieved results and future perspectives.

The paper presents a test method for evaluating air leakage in rodless cylinders with longitudinal slot and sealing band. To this aim a specific test rig was developed in order to carry out leakage tests under different operating conditions. Specifically, the performance of the sealing system consisting of the band, seal, and barrel is analyzed, identifying the critical conditions and factors that influence operation. Particular importance was assigned to selecting band and seal cross-sections in order to minimize leakage and thus limit energy consumption. A final rodless configuration with considerably reduced leakage was proposed.

The aim of secondary packaging plants is to pick food products from a conveyor belt and to place them into boxes. The typical configuration of these packaging plants consists of a set of sequential robot stations, performing pick and place cycles from one conveyor to another parallel one, which transport the products and the boxes to be filled. Depending on the relative movement of the two conveyors, the plant operates in co-current or counter-current flow configuration. Undesired perturbations in the product flow rate from its nominal value can lead to critical events, i.e. unpicked product at the end of the first conveyor or not-completely filled boxes. Even if the structures of co-current flow and of counter-current flow plants, are very similar, their behaviour in non-nominal or perturbed conditions can be significantly different. The aim of this paper is to deeply investigate the behaviour of these two kinds of secondary packaging lines, evaluating their performances in the case of different pick and place strategies, using discrete events simulation techniques. Results show to which extent the different proposed control strategies can improve the performances of both co-current and counter-currents plants and, in particular, how co-current plant layouts can achieve performances which are equivalent to, or perhaps even better than, those that can be obtained with a counter-current plant layout, that cannot be freely used since it has been patented. The simulation tool, control algorithms and results presented can help packaging plant designers for choosing the most appropriate solutions and for properly sizing the plant. Copyright © 2012 John Wiley & Sons, Ltd

The paper presents an experimental investigation of pneumatic resistances constructed using woven wire cloth. Tests were carried out on commercial Touraille stainless steel wire cloth featuring different mesh sizes. Two types of experiments were performed: (a) Mass flow rate through single resistances was measured with different upstream and downstream pressures and wire mesh sizes. (b) Pressure distribution and mass flow rate were measured on a pneumatic pad with a fine wire mesh feeding system. Two approaches of identification of the single resistances are discussed. Test results can be used as the basis for developing woven wire cloth pneumatic resistances whose conductance is equivalent to that of certain of the calibrated orifices normally used in gas bearings.

An experimental procedure is presented for measuring contact pressure between a pneumatic seal and its counterface. Measurements were performed using pressure-sensitive films, for which a suited calibration procedure was developed, also taking into account humidity influence. Having validated the experimental procedure on a lip seal rectilinear specimen, tests were carried out on commercial piston seals. Results were compared with those coming from numerical simulations performed using a finite-elements analysis.

In this paper, the problem of non minimal experiment design for the identification of static non linear systems affected by unknown but bounded errors is of concern. The memoryless non linear block is assumed to be a polynomial of finite and given order. Results are derived for the case in which the number of available measurements is larger than the number of parameters. Since, in this condition, the parameter uncertainties depend on the error realization and not only on the experiment design, some simulation results are presented to show the improvement that can be achieved with the proposed sampling schedules.

d -Sorbitol (SOR) is safe, is easy to measure, and has an exceptionally high extraction ratio in the normal liver of 0.93 ± 0.05 (mean ± SD). Together with the general interest in hepatic hemodynamics, these facts motivated us to review the usefulness of this compound for the assessment of liver plasma flow in humans. We concluded that in subjects without liver disease the nonrenal clearance of SOR—measured noninvasively—very closely approximates hepatic plasma flow. Because of its lower and more variable extraction ratio, indocyanine green should no longer be used without hepatic vein catheterization. Even in patients with cirrhosis, SOR exhibits higher hepatic extraction ratios than indocyanine green. To fully explore the potential of SOR in the evaluation of such patients attention needs to be paid to the complex changes in architecture and function occurring in this disease. In cirrhotics the noninvasively measured nonrenal clearance of SOR presumably approximates the flow through intact and capillarized sinusoids (functional flow) and reflects the amount of blood having functional contact with hepatocytes. The theoretic background of the method, its accuracy, further research needs, and potentials of various approaches are discussed in detail.

In this paper consequences of low bit representation and pling on parameter identification are considered. Low bit representation that can be motivated by speed and cost constraints on A/D converters, induces bounded measurement errors on data. Such errors appear to be uniformly distributed so that parameter identification with algorithms developed in the bounded noise context is effective. Possible additional gaussian measurement errors does not affect the performances especially when their variance is low compared with the truncation error. If oversampling is possible, preprocessing of oversampled data with central algorithms is effective and leads to final identification results similar to those that could be achieved with higher bit representation. The different features are illustrated with a simulation study on a MA system.

In food processing plants, raw materials are fed into the system in different supply-lots of product, and are processed through different stages. In these stages, raw or intermediate materials are mixed or combined together, and physico-chemical and/or microbiological processes such as heating, concentration, pasteurisation etc. take place. In this setting, traceability consists of the ability to determine for each portion of intermediate or final product, in any part of the plant, its relative composition in terms of supply-lots fed into the system as well as of new lots generated during the production process.Traceability becomes particularly difficult in the very common case when bulk products, such as liquids or grains, are involved in the production chain. Current traceability practices are in most cases unable to directly deal with bulk products, and typically resort to the definition of very large lots to compensate the lack of knowledge about lot composition. As demonstrated in recent food crises, this over-bounding approach has weaknesses in clearly identifying, immediately after risk assessment, the affected product lots, leading to unavoidably wide, expensive and highly impacting recalls.Motivated by these considerations, this paper presents a novel approach to manage traceability of bulk products during production, storage and delivery. It provides a tight definition of lots in terms of their composition and size, thus allowing strict control of the production and supply chains. © 2013 IAgrE.

A new artifact-based method to identify the systematic errors in multi-axis CNC machine tools minimizing the worst case prediction error is presented. The closed loop volumetric error is identified by simultaneously moving the axes of the machine tool. The physical artifact is manufactured on the machine tool and later measured on a coordinate measuring machine. The artifact consists of a set of holes in the machine tool workspace at locations that minimize the worst case prediction error for a given bounded measurement error. The number of holes to be drilled depends on the degree of the polynomials used to model the systematic error and the number of axes of the machine tool. The prediction error is also function of the number and location of the holes. The feasibility of the method is first investigated for a two-axis machine to find the best experimental setting. Finally based on the two-axis case study, we extend the results to machine tools with any number of axes. The obtained results are very promising and require only a short time to produce the artifact

In this paper, we examine optimal sequences that generate worst-case parameters estimation errors in the l 1 , l 2 and l ∞ norm context for algorithms identifying linear, time-invariant discrete-time, finite impulse response systems excited by bounded sequences and with l ∞ norm measurement error.

The aim of this paper is to introduce the GnssEduPack, which is a training kit in the field of Global Navigation Satellite System (GNSS) which is oriented towards university students and GNSS application developers. The whole idea behind the GnssEduPack is to provide students with a new tool, which, beside providing theoretical background and explanations, allows for consistent experimental activity with real GNSS signals thus bringing the student directly in contact with practice. The key feature of the GnssEduPack is the emphasis on practical problems of real GNSS oriented engineering, which is made possible by the use of the SAT_SURF Suite, a light, compact and easy-to-use hardware/software experimental platform.

In this paper, the problem of parameter identification for models with bounded measurement errors both on the input and on the output is addressed and some corrections to previously published results are presented. In particular, it is shown that only parameter overbounds can in general be computed for systems of the form y = (φ + δφ)θ + δy when the bounded measurement errors δφ and δy are correlated. Since ARMAX and bilinear systems can be represented in this form, it turns out that tight parameter bounds are in general not available for these systems. Finally, we show that it is possible to check a posteriori whether the obtained bounds are tight or not.

In this paper attention is restricted to linear systems described by y = Ar + e where the measurement error vector is unknown but bounded. In this context, the behaviour ot two new recursive algorithms for the central and projection estimates determination is investigated. Λ simulation study is performed on some significant examples and the obtained results are compared with those of other off-line algorithms.

This paper discusses the problem of parameter identification in presence of errors and uncertainties characterized by a set-membership description. Some relations existing between posterior identifiability, structure selection and validation, and the influence of modeling and measurement error are pointed out. Both theoretical and algorithmic problems are discussed; special attention is devoted to the influence of various assumptions and a priori information on the algorithmic complexity. Finally a practical example, derived from the tuning procedure of a digital voltmeter in an intelligent measurement equipment, is presented.

Despite the large diffusion of robotic and automated solutions that took place during the last decades in most production processes, the agricultural sector only marginally benefited from automated solutions (such as the control of climatic parameters in greenhouses). Robotic applications have been confined so far almost only to research studies with few particular and specific applications made available for farmers. This lack of advanced robotic solutions in agriculture and particularly in intensive farming cannot be motivated claiming a lack of interest for the latest results of technology and research in this sector. The use of latest results of genetics for the production of hybrid flower plants is one of the many examples that contradicts this kind of argument and shows how much the agricultural sector is keen to exploit the opportunities offered by modern technology and research. However, from the automation point of view, agriculture remains mainly labour intensive not only in those countries where manpower is relatively cheap, but also for those enterprises which enthusiastically embrace latest technologies in an effort to improve their competitiveness and to ensure top quality products. The motivation for the little development that robotic solutions deserved so far in the agricultural sector lies elsewhere. On one side it is related to some particularities of the specific sector, like the fact that farming is usually performed in an unstructured environment that is therefore less friendly for robotic solutions than a well structured industrial environment (Kassler, 2001). On the other side a consistent share of the research effort conducted so far, mainly tried to use standard industrial robotic solutions adapting them to the intensive farming sector instead of developing brand new solutions that exploit the specific features of the agricultural sector. Finally the focus of most research was so far on single specific activities or tasks and less frequently on the whole production process or on a consistent share of it. In our opinion the approach should be revised looking for new specific robotic solutions that take advantage of peculiarities and needs encountered in the agricultural sector that are different from those of the industry. In this context, beside overcoming specific constraints encountered in the farming context, the important challenge is to find solutions and develop technology to automatize consistent shares of the agricultural process. From a business point of view it is important to stress that, for the

In this paper the problem of optimal input design for the identification of Hammerstein models is considered under the assumption that the linear dynamic part of the model is a FIR and that lower and upper bounds are available for the additive measurement errors. The parameters of the Hammerstein model can then be estimated via the identification of a linearized augmented Hammerstein model . External approximations of the feasible intervals for the parameters of the original Hammerstein models are then derived (which may correspond to the actual feasible intervals). This paper deals with the design of input sequences minimizing parameter uncertainty for the linearized augmented Hammerstein model . Some new results are also reported about optimal input design for polynomial non-linear blocks, that may be part of Hammerstein models.

This chapter reviews some of the more recent algorithms for sequential parameter identification in the context of unknown but bounded measurement errors when the model output is linear in the parameters. The properties of the different algorithms are analyzed and compared.

In this paper we consider parameter estimation of linear systems described by y i = a i T θ + e i , where the ith measurement y i is linearly dependent on the parameter vector θ ∈ R p through the regressor vector a i T ∈ R p and the measurement error e i is unknown but bounded. Some properties of previously presented algorithms for recursive parameter identification in the unknown but bounded error (UBBE) context are discussed. In particular it is analysed how different levels of information on the error structure can influence the choice of the identification algorithms and the possibility of evaluating the reliability of the estimates. Attention is also focused on the influence that forgetting schemes have on the estimates and on their confidence evaluation

Two algorithms for sequential parameter identification when the measurement errors are not statistically described are introduced. These are the projection estimate algorithm and the central estimate algorithm. Their convergence properties are illustrated, and a comparison with existing algorithms is performed. In practical applications it seems that the best choice for sequential parameter identification is the algorithm that gives the projection estimate. It is computationally lighter than the algorithm that computes the central estimate, and requires only the knowledge of the relative weights of the measurement errors and not their actual values. >

The problem of parameter identification from measurements corrupted by unknown-but-bounded errors is addressed. It mainly consists in finding the minimal parameter set ∧ consistent both with the obtained data samples and the assumed error description. This problem was treated by two different approaches; one using an ellypsoidal upper bound on ∧ via a recursive algorithm, while in the other approach an hypercubic bound was obtained through the solution of a linear programming problem, The hypercubic description gives an exact bound to the parameter uncertainty, consistent with the available measurements and the error model, but is time consuming when a large number of measurements is present. The ellypsoidal approach allows a fast data preprocessing, producing a smaller subset of constraints that can be fed to the linear programming algorithm. In the present paper a comparison between the two approaches is outlined and a combined use of the positive aspects of both procedures is proposed. The original algorithm has been modified for higher efficiency and the results are compared and discussed.

A system has been devised for the fast assembly of pegs in holes using a guiding frame which has four degress of freedom and which is separated from the insertion actuator.

The problem of optimal sampling design for parameter estimation when data are generated by linear models is addressed. The measurements are assumed to be corrupted by an unknown but bounded additive noise. The sampling design assumes that the number of samples is unconstrained and no replication is allowed. Two main results are shown: 1) for particular classes of linear models, the optimal number of measurements is equal to the number of parameters, as in the statistical context; 2) the uncertainty intervals of the parameter estimates are bounded from above by quantities that can be computer a priori, knowing only the model and the error structure.

Measurements are inevitably bound to experimental errors, whose probability distributions are generally unknown. When models of metabolic-endocrine systems or of drug kinetics are identified starting from noisy data, the question arises of how the parameters estimates are affected by the assumptions made. Aim of this work is to investigate whether and to what extent unreliable assumptions concerning the noise distribution influence the estimation process and to compare the efficiency of two estimation methods in the case of ‘ad hoc’ simulated data. Al though these are generated from two-compartment systems which can model some aspects of a β-blocking agent kinetics, both linear and nonlinear, it is believed that the findings of this study are of more general interest. The results show that the minimizing algorithms do not significantly differ as for their capability of fitting a bi-exponential curve to the simulated data. Evidence is given of the ambiguities encountered when hypotheses concerning the system structure are tested against the algorithms results. Most of the findings of this work can be expected from estimation theory; however they allow a better insight into the limits of applying estimation methods to biomedical problems, because the systems structure has been drawn from drug kinetics and a large variety of additive noise characteristics has been considered.

The effect of three isolated defects in the enterohepatic circulation of bile acids on the size and distribution of the bile acid pool, plasma bile acid levels and bile acid secretion into the intestine was simulated using a linear multicompartmental physiological pharmacokinetic model previously used to simulate these aspects of bile acid metabolism in healthy man. Stepwise increases in portal-systemic shunting (with a reciprocal decrease in hepatic blood flow) caused an exponential increase in systemic plasma concentrations of bile acids, but no other major changes in bile acid metabolism. When the effect of varying fractional hepatic extraction was simulated, it was found that the greater the fractional hepatic extraction, the greater the elevation observed for systemic plasma bile acid levels for a given degree of portal-systemic shunting. When total hepatic blood flow was restored to normal by simulating "arterialization," systemic plasma levels of bile acids decreased strikingly, yet remained elevated. For cholate with a fractional hepatic extraction of 0.9 and 100% portal-systemic shunting, arterialization caused a decrease from a 20-fold elevation to a 5-fold elevation. This simulation thus defined the effect of the presence of the portal venous system per se on plasma bile acid levels and also quantified the circulatory route by which substances reach the liver when portal-systemic shunting is present. An isolated defect in hepatic uptake of bile acids caused little change in overall bile acid metabolism other than modestly increased plasma levels. Loss of bile acid storage by the gallbladder caused the majority of the bile acid pool to move from the gallbladder compartments to the proximal small intestine during fasting but had little effect on the dynamics of the enterohepatic circulation during eating. The results of these novel simulations of isolated defects in bile acid transport should aid in the interpretation of the more complex changes in bile acid metabolism which are likely to occur in hepatic or biliary disease.

The problem of evaluating upper and lower bounds of the parameters of dynamic systems is considered when a probabilistic description of noise and errors is not available (or suitable), but a bound on them is known. Two slightly different problems are defined (evaluation of bounds of least mean square estimates and of the parameters themselves) and simple and computationally feasible solutions are proposed.

A multicompartmental pharmacokinetic model based on physiological principles, experimental data, and the standard mathematical principles of compartmental analysis has been constructed that fully describes the metabolism and enterohepatic cycling in man of cholic acid, a major bile acid. The model features compartments and linear transfer coefficients. The compartments are aggregated into nine spaces based on physiological considerations (liver, gallbladder, bile ducts, jejunum, ileum, colon, portal blood sinusoidal blood, and general circulation). The transfer coefficients are also categorized according to function: flow, i.e., emptying of gallbladder or intestinal spaces, and circulation of the blood; biotransformation, i.e., conjugation, deconjugation, or dehydroxylation; and transport, i.e., active or passive transport. The model is made time dependent by introducing meals, which trigger discrete increases in gallbladder emptying and intestinal flow. Each space contains three compartments. For cholic acid, these are unconjugated cholic acid, cholylglycine, and cholyltaurine. The model was then used with all existing experimental data to simulate cholic acid metabolism in healthy man over a 24-h period. Satisfactory agreement was obtained between simulated and experimental results for serum bile acid levels, hepatic bile acid secretion, and bile acid secretion into the intestine. The model was also used to classify 16 clinical instances in which the enterohepatic circulation of bile acids is altered by drugs or disease. The model can be extended to describe completely the metabolism and enterohepatic circulation of any bile acids in man in health and digestive disease. The model should also be broadly applicable to the description of the pharmacokinetics of all other drugs whose metabolism is similar to that of bile acids, i.e., drugs for which there are tissue and bacterial biotransformations, enterohepatic cycling, and appreciable first-pass clearance.

In this paper we address the problem of parameter estimation of a linear model y = A λ + ρ where the input matrix A is known and the additive uncertainty ρ is assumed to be unknown but bounded in an l ∞ norm by a given constant ∊. In this case for given data y the set of all admissible parameters λ consistent with the given model equation and the bound ϵ is a convex polytope. In this paper we attempt to estimate optimally, in the sense of maximizing the volume, a rectangular box which is guaranteed to lie completely inside the true region. The inner bounding region obtained in this way can be used together with an outer bounding region, in order to approximate as closely as possible the boundary of the actual admissible parameter polytope. A computable algorithm, using linear programming, is proposed and some numerical results are given to illustrate the behavior of the technique.

The problem of parameter estimation, when the relation between parameters and measurements is nonlinear, is addressed in the context of set membership uncertainty; it means that the additive measurement error is considered to be an element of a given set, no statistical assumption on its distribution being made. Unlike the linear case where exact solutions can be computed, in the nonlinear case it appears that only approximate results can be achieved in general. This fact is related to the many different unexpected features that the nonlinear relation can introduce. Most of these features are illustrated comparing the linear and nonlinear case, and they are shown by means of some very simple examples. The available algorithms are then presented and their limitations are discussed

A syntactic method for analysing ECG waveforms considered as a multichannel process is presented. A procedure is designed in order to extract the time evolution of the rhythm using the energy of the ECG derivatives.

The problem of the best use of experimental data for qualitative and quantitative evaluation of liver disfunction is analysed. Only statistical procedures for classification are considered. Emphasis is placed on a complete description of the most important steps that must be performed (degree of clustering of the data, validity of the hypotheses underlying the statistical methods used, meaning and adequate use of the reference classification, reliability of the different statistical methods used, information content of the laboratory tests studied) since the omission of any of them may contribute to a misleading interpretation of the overall results or may prevent significant practical utilization. An application to BSP tests is carried out: results obtained from 350 subjects are presented and discussed. A classification related to the main functional aspects of liver impairment was chosen. Different combinations of the parameters obtained from BSP blood disappearance curve — initial disappearance rate (K 1 ), 45 min-retention of total (RT45) and conjugated (RC45) BSP- and different mathematical algorithms are compared. Five classes (N, normal subjects; PN, paranormal subjects; H, prevalent hepatocellular damage; C, prevalent cholestatic damage; HC, heavy combined damages) can be satisfactorily discriminated with all three parameters (K 1 + RT45 + RC45); however K 1 + RC45 give almost the same results and represent an interesting compromise between data information content and laboratory complexity.

When a bayesian classification algorithm has to be performed on a set of continuous variables, their probability density functions (pdfs) have to be known for each one of the pattern classes.

Bile acids are important endogenous compounds involved in many pathophysiological processes. The synthetic production of natural bile acids, as well as that of new ones not present in nature, is of increasing interest as those compounds are becoming used as drugs. Among the novel bile acids, nor-deoxycholate (nor-DC) presents an unespected high choleretic effect that, in the authors opinion, could be related to a cholehepatic recirculation. A modeling approach to determine the presence or the absence of the cholehepatic pathway is presented together with the first numerical results obtained from the available experimental data. From these results it appears that the cholehepatic recirculation of nor-DC is likely to occur; however further investigation, with an optimization of the experimental design, are recommended.

Since Sir Ronald Fisher (1) proposed in 1935 a discriminant function analysis with an application to a taxonomic problem,many methods for classifying subjects into one of a set of classes, given a series of measurements made on each subject, have been extensively developed.

When solving parameter estimation problems, it is common to assume more or less implicitly that the measurement errors are normally distributed; suitable estimates are then computed as well as their covariance matrix that represents the reliability of the estimates themselves. Indeed, there are many applications where measurement and a priori information are too sparse to provide adequate error statistics, and, in other cases, the measurement error is known to be — at least in part — not stochastic. To deal with such problems in the last years, new approaches have been proposed that refer to a bounded error model; in other words, only an interval of variability of the error is defined. This paper reviews some former results, introduces a geometrical representation. and establishes a link between statistical assumptions and set error description.

A computer program for sequential bayesian classification of patterns defined by integer and real-valued data is described. Classified patterns from a training sample are used to estimate the non-parametric (kernel) probability density functions and the a-priori class probabilities necessary to implement the bayesian classification. For each pattern and at each step in the sequential program, the 'best' feature to be measured at the next step is computed on the basis of the estimated misallocation error rate. The user can actually use the proposed feature or any other one; once the chosen feature has been measured, its value is used to allocate the pattern into the class with the highest conditional a-posteriori probability, according to the Bayes formula. The main feature of the program consists in the computation of the 'probability of reversal' at each step of the sequential procedure. The probability of reversal represents the probability that at the next step the pattern will be classified into a class different from the present one. The probability of reversal can be used as a stopping criterion, which is more efficient than other commonly used stopping rules, such as the a-posteriori Bayes probability or the estimated misallocation error rate. The program, available in FORTRAN 77 for a VAX/VMS machine, has been tested both on simulated and real data collected from patients suffering from various forms of hepatic disease.

The problem of parameter estimation and of the evaluation of related uncertainty intervals is considered in the case that a probabilistic description of noise and errors is not available (of suitable), but only a bound on them is known. In the present paper only linear parametrizations and estimators are considered. Very simple and computationally feasible algorithms are derived for evaluating two different types of uncertainty intervals (Estimates Uncertainty Intervals, Parameter Uncertainty Intervals). The relationships between the EUI's and the PUI's are established, and the solution to the problem of the minimum uncertainty intervals estimator is given: the latter can be obtained by means of a simple linear programming algorithm.