To Örebro University

In this dissertation, we present and evaluate algorithms for statistical gas distribution modelling in mobile robot applications. We derive a representation of the gas distribution in natural environments using gas measurements collected with mobile robots. The algorithms fuse different sensors readings (gas, wind and location) to create 2D or 3D maps.

Throughout this thesis, the Kernel DM+V algorithm plays a central role in modelling the gas distribution. The key idea is the spatial extrapolation of the gas measurement using a Gaussian kernel. The algorithm produces four maps: the weight map shows the density of the measurements; the confidence map shows areas in which the model is considered being trustful; the mean map represents the modelled gas distribution; the variance map represents the spatial structure of the variance of the mean estimate.

The Kernel DM+V/W algorithm incorporates wind measurements in the computation of the models by modifying the shape of the Gaussian kernel according to the local wind direction and magnitude.

The Kernel 3D-DM+V/W algorithm extends the previous algorithm to the third dimension using a tri-variate Gaussian kernel.

Ground-truth evaluation is a critical issue for gas distribution modelling with mobile platforms. We propose two methods to evaluate gas distribution models. Firstly, we create a ground-truth gas distribution using a simulation environment, and we compare the models with this ground-truth gas distribution. Secondly, considering that a good model should explain the measurements and accurately predicts new ones, we evaluate the models according to their ability in inferring unseen gas concentrations.

We evaluate the algorithms carrying out experiments in different environments. We start with a simulated environment and we end in urban applications, in which we integrated gas sensors on robots designed for urban hygiene. We found that typically the models that comprise wind information outperform the models that do not include the wind data.

Environmental monitoring is a rather new field in robotics. One of the main appealing tasks is gas mapping, i.e., the characterization of the chemical properties (concentration, dispersion, etc.) of the air within an environment. Current approaches rely on a robot using standard localization and mapping techniques to fuse gas measures with spatial features. These approaches require sophisticated sensors and/or high computational resources. We propose a minimalistic approach, in which one or multiple low-cost robots exploit the ability to store information in the environment, or “stigmergy”, to effectively compute an artificial potential leading toward the likely location of the gas source, as indicated by a highest gas concentration or fluctuation. The potential is computed and stored directly on an array of RFID tags buried under the floor. Our approach has been validated in extensive experiments performed on real robots in a domestic environment.

Gas distribution models can provide comprehensive information about a large number of gas concentration measurements, highlighting, for example, areas of unusual gas accumulation. They can also help to locate gas sources and to plan where future measurements should be carried out. Current physical modeling methods, however, are computationally expensive and not applicable for real world scenarios with real-time and high resolution demands. This chapter reviews kernel methodsthat statistically model gas distribution. Gas measurements are treated as randomvariables, and the gas distribution is predicted at unseen locations either using akernel density estimation or a kernel regression approach. The resulting statistical 

apmodelsdo not make strong assumptions about the functional form of the gas distribution,such as the number or locations of gas sources, for example. The majorfocus of this chapter is on two-dimensional models that provide estimates for themeans and predictive variances of the distribution. Furthermore, three extensionsto the presented kernel density estimation algorithm are described, which allow toinclude wind information, to extend the model to three dimensions, and to reflecttime-dependent changes of the random process that generates the gas distributionmeasurements. All methods are discussed based on experimental validation usingreal sensor data.

In the framework of DustBot European project, aimed at developing a new multi-robot system for urban hygiene management, we have developed a twowheeled robot: DustCart. DustCart aims at providing a solution to door-to-door garbage collection: the robot, called by a user, navigates autonomously to his/her house; collects the garbage from the user and discharges it in an apposite area. An additional feature of DustCart is the capability to monitor the air pollution by means of an on board Air Monitoring Module (AMM). The AMM integrates sensors to monitor several atmospheric pollutants, such as carbon monoxide (CO), particular matter (PM10), nitrogen dioxide (NO2), ozone (O3) plus temperature (T) and relative humidity (rHu). An Ambient Intelligence platform (AmI) manages the robots’ operations through a wireless connection. AmI is able to collect measurements taken by different robots and to process them to create a pollution distribution map. In this paper we describe the DustCart robot system, focusing on the AMM and on the process of creating the pollutant distribution maps. We report results of experiments of one DustCart robot moving in urban scenarios and producing gas distribution maps using the Kernel DM+V algorithm. These experiments can be considered as one of the first attempts to use robots as mobile monitoring devices that can complement the traditional fixed stations.

We undertook a meta-analysis of six Crohn's disease genome-wide association studies (GWAS) comprising 6,333 affected individuals (cases) and 15,056 controls and followed up the top association signals in 15,694 cases, 14,026 controls and 414 parent-offspring trios. We identified 30 new susceptibility loci meeting genome-wide significance (P < 5 × 10⁻⁸). A series of in silico analyses highlighted particular genes within these loci and, together with manual curation, implicated functionally interesting candidate genes including SMAD3, ERAP2, IL10, IL2RA, TYK2, FUT2, DNMT3A, DENND1B, BACH2 and TAGAP. Combined with previously confirmed loci, these results identify 71 distinct loci with genome-wide significant evidence for association with Crohn's disease.

In this paper we present a statistical method to build three-dimensional gas distribution maps from gas sensor and wind measurements obtained with a mobile robot in uncontrolled environments. The particular contribution of this paper is to introduce and evaluate an algorithm for 3D statistical gas distribution mapping, that takes into account airflow information. 3D-Kernel DM+V/W algorithm uses a multivariate Gaussian weighting function to model the information provided by the gas sensors and an ultrasonic anemometer. The proposed algorithm is evaluated with respect to the ability of the obtained models to predict unseen measurements. The results based on 15 trials with a mobile robot in an indoor environment show improvements in the model performance when using the 3D kernel DM+V/W algorithm. Moreover the model is able to adapt to the dynamical changes of the environment learning the hyper-parameter from the sensors readings.

The EU project DustBot addresses urban hydeience. Two types of robots were designed, the DustClean robot to autonomously clean pedestrian areas, and the DustCart robot for door-to-door garbage collection. Three prototype robots were built and equipped with electronic noses so as to enable them to collect environmental data while performing their urban hygiene tasks. Essentially, the robots act as a mobile, wirless node in a sensor network. In this paper we give an overview of the DusBot platform focusig on the Air Monitoring Module (AMM). We descibe the data flow between the robots throught the ubiquitous network to a gas distribution modelling server, where a gas deisribution model is computed. We descibe the Kernel DM+V algorithn, an approach to create statistical gas disdtribution models in the form of predictive mean and variance discrtized onto a grid map. Finally we present and discuss results obtained with the DustBot AMM during experimental trails performex in outdoor public places; a courtyard in Pontedera, Italy and a pedestrian square in Örebro, Sweden.

Gas distribution modelling constitutes an ideal application area for mobile robots, which – as intelligent mobile gas sensors – offer several advantages compared to stationary sensor networks. In this paper we propose the Kernel DM+V algorithm to learn a statistical 2-d gas distribution model from a sequence of localized gas sensor measurements. The algorithm does not make strong assumptions about the sensing locations and can thus be applied on a mobile robot that is not primarily used for gas distribution monitoring, and also in the case of stationary measurements. Kernel DM+V treats distribution modelling as a density estimation problem. In contrast to most previous approaches, it models the variance in addition to the distribution mean. Estimating the predictive variance entails a significant improvement for gas distribution modelling since it allows to evaluate the model quality in terms of the data likelihood. This offers a solution to the problem of ground truth evaluation, which has always been a critical issue for gas distribution modelling. Estimating the predictive variance also provides the means to learn meta parameters and to suggest new measurement locations based on the current model. We derive the Kernel DM+V algorithm and present a method for learning the hyper-parameters. Based on real world data collected with a mobile robot we demonstrate the consistency of the obtained maps and present a quantitative comparison, in terms of the data likelihood of unseen samples, with an alternative approach that estimates the predictive variance.

Recent publications in statistical gas distribution modelling have proposed algorithms that model mean and variance of a distribution. This paper argues that estimating the predictive concentration variance entails not only a gradual improvement but is rather a significant step to advance the field. This is, first, since the models much better fit the particular structure of gas distributions, which exhibit strong fluctuations with considerable spatial variations as a result of the intermittent character of gas dispersal. Second, because estimating the predictive variance allows to evaluate the model quality in terms of the data likelihood. This offers a solution to the problem of ground truth evaluation, which has always been a critical issue for gas distribution modelling. It also enables solid comparisons of different modelling approaches, and provides the means to learn meta parameters of the model, to determine when the model should be updated or re-initialised, or to suggest new measurement locations based on the current model. We also point out directions of related ongoing or potential future research work.

In this paper we present a statistical evaluation of the Kernel DM+V/W algorithm to build two-dimensional gas distribution maps with a mobile robot. In addition to gas sensor measurements from an "e-nose" the Kernel DM+V/W algorithm also takes into account wind information received from an ultrasonic anemometer. We evaluate the method based on real measurements in three uncontrolled environments with very different properties. As a measure for the model quality we compute how well unseen measurements are predicted in terms of the data likelihood. A paired Wilcoxon signed rank test shows a significant improvement (at a confidence level of 95%) of the model quality when using wind information.