Artificial Olfactory System Research Articles

Shuffled Frog-Leaping and Weighted Cosine Similarity for Drift Correction in Gas Sensors

Artificial Olfactory Systems (AOS) mimic the Biological Olfaction (BO) using sensors and artificial intelligence algorithms. The performance of an AOS is based on the sensitivity and the selectivity of the sensors against the sensed odors. Long term sensors drift is a major problem that brings distortion in the sensitivity of the sensors. Traditional methods to compensate drift, like sensors replacement, successive recalibration or domain transformations are either expensive or not always feasible to use. In order to overcome the issues related to traditional approaches, a novel method based on a sequential classification approach is proposed in this paper, with following contributions: (i) A Tree Structured Cosine Similarity (TSCS) based classification approach is proposed to handle long term sensors drift. (ii) The classifier is embedded within a memetic metaheuristic Shuffled Frog Leaping Optimization (SFLO) approach to optimize the features space. (iii) The proposed approach works as a combinatorial optimization problem that recursively reduces the number of features and increases the classification accuracy of the system, while compensating the drift. (iv) Only median values of optimized features are enough to train the classifier which reduces the computational cost and the memory requirement of the classifier. (v) The proposed approach is purely based on the feature subset selection process for drift compensation without requiring any data samples from target domain or system recalibration, making it suitable for the real life applications. The proposed approach is compared with the existing state-of-the-art approaches using an extensive experimental dataset and a significant increase in the classification accuracy is observed.

Development of artificial olfactory system

Development of artificial olfactory system

A Hardware-Deployable Neuromorphic Solution for Encoding and Classification of Electronic Nose Data.

In several application domains, electronic nose systems employing conventional data processing approaches incur substantial power and computational costs and limitations, such as significant latency and poor accuracy for classification. Recent developments in spike-based bio-inspired approaches have delivered solutions for the highly accurate classification of multivariate sensor data with minimized computational and power requirements. Although these methods have addressed issues related to efficient data processing and classification accuracy, other areas, such as reducing the processing latency to support real-time application and deploying spike-based solutions on supported hardware, have yet to be studied in detail. Through this investigation, we proposed a spiking neural network (SNN)-based classifier, implemented in a chip-emulation-based development environment, that can be seamlessly deployed on a neuromorphic system-on-a-chip (NSoC). Under three different scenarios of increasing complexity, the SNN was determined to be able to classify real-valued sensor data with greater than 90% accuracy and with a maximum latency of 3 s on the software-based platform. Highlights of this work included the design and implementation of a novel encoder for artificial olfactory systems, implementation of unsupervised spike-timing-dependent plasticity (STDP) for learning, and a foundational study on early classification capability using the SNN-based classifier.

Smart gas sensor arrays powered by artificial intelligence

Mobile robots behaving as humans should possess multifunctional flexible sensing systems including vision, hearing, touch, smell, and taste. A gas sensor array (GSA), also known as electronic nose, is a possible solution for a robotic olfactory system that can detect and discriminate a wide variety of gas molecules. Artificial intelligence (AI) applied to an electronic nose involves a diverse set of machine learning algorithms which can generate a smell print by analyzing the signal pattern from the GSA. A combination of GSA and AI algorithms can empower intelligent robots with great capabilities in many areas such as environmental monitoring, gas leakage detection, food and beverage production and storage, and especially disease diagnosis through detection of different types and concentrations of target gases with the advantages of portability, low-power-consumption and ease-of-operation. It is exciting to envisage robots equipped with a "nose" acting as family doctor who will guard every family member's health and keep their home safe. In this review, we give a summary of the state-of the-art research progress in the fabrication techniques for GSAs and typical algorithms employed in artificial olfactory systems, exploring their potential applications in disease diagnosis, environmental monitoring, and explosive detection. We also discuss the key limitations of gas sensor units and their possible solutions. Finally, we present the outlook of GSAs over the horizon of smart homes and cities.

Performance analysis of gas sensing device and corresponding IoT framework in mines

Gas sensor devices usage have found prominence in fields like artificial olfactory systems and is often used to keep in check pollution and other toxic gas hazards. Graphene mono-layered sensing material along with Gold nano-particles have been previously used to formulate a sensor device. In recent times, internet of things (IoT) capable gas sensing device is a subject of great interest. Previously, it has been implemented through various wired and wireless communications like Zigbee Protocol, IBM Bluemix and sensor nodes but they have their own limitations. In this work nano gas sensor device used for detecting perilous and toxic gas has been fabricated and enhanced to increase its selectivity and sensitivity. Fabrication of Au based TiO2, CuO, ZnO, and WO3 resistive type gas sensor array has been done using sol gel method and analyzed to ensure better conduction. Application of the proposed sensing device can prevent accident in mines by precise and prompt checking of hazardous gases present inside. An IoT frame work has been proposed in this study where Node MCU (ESP8266) is used as a WiFi module along with Message Queuing Telemetry Transport protocol, which resulted in better speed and efficiency. Increased sensitivity and selectivity of the sensor nodes with better efficiency and speed of communication ensures a productive mine hazard management.

Optimization of Modulation Methods for Solenoid Valves to Realize an Odor Generation System.

An artificial olfactory system coupled with an odor generation system is herein reported. The artificial olfactory system is composed of four chemical sensors consisting of quartz crystal microbalances (QCMs) coated with room temperature ionic liquids (RTILs). The sensors are interrogated by four vector network analyzers, which are used to measure the series resonant frequency and motional resistance. The odor generation system can generate eight different odors and mix them in any composition. Solenoid valves are used to switch the path and control the concentration of the different odors before blending. Two algorithms to control the solenoid valves, delta-sigma modulator, and simple pulse width modulation (PWM) are studied, optimized, and compared. Finally, the uncertainty of the odor generating system is calculated.

Surface-Enhanced Raman Scattering-Based Odor Compass: Locating Multiple Chemical Sources and Pathogens.

Olfaction is important for identifying and avoiding toxic substances in living systems. Many efforts have been made to realize artificial olfaction systems that reflect the capacity of biological systems. A sophisticated example of an artificial olfaction device is the odor compass which uses chemical sensor data to identify odor source direction. Successful odor compass designs often rely on plume-based detection and mobile robots, where active, mechanical motion of the sensor platform is employed. Passive, diffusion-based odor compasses remain elusive as detection of low analyte concentrations and quantification of small concentration gradients from within the sensor platform are necessary. Further, simultaneously identifying multiple odor sources using an odor compass remains an ongoing challenge, especially for similar analytes. Here, we show that surface-enhanced Raman scattering (SERS) sensors overcome these challenges, and we present the first SERS odor compass. Using a grid array of SERS sensors, machine learning analysis enables reliable identification of multiple odor sources arising from diffusion of analytes from one or two localized sources. Specifically, convolutional neural network and support vector machine classifier models achieve over 90% accuracy for a multiple odor source problem. This system is then used to identify the location of an Escherichia coli biofilm via its complex signature of volatile organic compounds. Thus, the fabricated SERS chemical sensors have the needed limit of detection and quantification for diffusion-based odor compasses. Solving the multiple odor source problem with a passive platform opens a path toward an Internet of things approach to monitor toxic gases and indoor pathogens.

A Novel Method for Soil Organic Matter Determination by Using an Artificial Olfactory System.

Soil organic matter (SOM) is a major indicator of soil fertility and nutrients. In this study, a soil organic matter measuring method based on an artificial olfactory system (AOS) was designed. An array composed of 10 identical gas sensors controlled at different temperatures was used to collect soil gases. From the response curve of each sensor, four features were extracted (maximum value, mean differential coefficient value, response area value, and the transient value at the 20th second). Then, soil organic matter regression prediction models were built based on back-propagation neural network (BPNN), support vector regression (SVR), and partial least squares regression (PLSR). The prediction performance of each model was evaluated using the coefficient of determination (R2), root-mean-square error (RMSE), and the ratio of performance to deviation (RPD). It was found that the R2 values between prediction (from BPNN, SVR, and PLSR) and observation were 0.880, 0.895, and 0.808. RMSEs were 14.916, 14.094, and 18.890, and RPDs were 2.837, 3.003, and 2.240, respectively. SVR had higher prediction ability than BPNN and PLSR and can be used to accurately predict organic matter contents. Thus, our findings offer brand new methods for predicting SOM.

Neural Network Technique for Electronic Nose Based on High Sensitivity Sensors Array

Electronic Nose, as an artificial olfaction system, has potential applications in environmental monitoring because of its proven ability to recognize and discriminate between a variety of different gases and odors. In this paper, we used a chemical sensor array to develop an electronic nose to detect and identify seven different gases (H2, C2H2, CH4, CH3OCH3, CO, NO2, and NH3). These gas sensors are chosen because of its hierarchical/doped nanostructure characteristics, which give them a very high sensitivity and low response time; we improve the linearity response and temperature dependence using models based on artificial neural networks. We used in Electronic nose a pattern recognition based on artificial neural network, which discriminates qualitatively and quantitatively seven gases and has a fast response.

Real-Time Classification of Multivariate Olfaction Data Using Spiking Neural Networks.

Recent studies in bioinspired artificial olfaction, especially those detailing the application of spike-based neuromorphic methods, have led to promising developments towards overcoming the limitations of traditional approaches, such as complexity in handling multivariate data, computational and power requirements, poor accuracy, and substantial delay for processing and classification of odors. Rank-order-based olfactory systems provide an interesting approach for detection of target gases by encoding multi-variate data generated by artificial olfactory systems into temporal signatures. However, the utilization of traditional pattern-matching methods and unpredictable shuffling of spikes in the rank-order impedes the performance of the system. In this paper, we present an SNN-based solution for the classification of rank-order spiking patterns to provide continuous recognition results in real-time. The SNN classifier is deployed on a neuromorphic hardware system that enables massively parallel and low-power processing on incoming rank-order patterns. Offline learning is used to store the reference rank-order patterns, and an inbuilt nearest neighbor classification logic is applied by the neurons to provide recognition results. The proposed system was evaluated using two different datasets including rank-order spiking data from previously established olfactory systems. The continuous classification that was achieved required a maximum of 12.82% of the total pattern frame to provide 96.5% accuracy in identifying corresponding target gases. Recognition results were obtained at a nominal processing latency of 16ms for each incoming spike. In addition to the clear advantages in terms of real-time operation and robustness to inconsistent rank-orders, the SNN classifier can also detect anomalies in rank-order patterns arising due to drift in sensing arrays.

Electronic nose for smart identification of roofing and paving grade asphalt

Abstract Asphalt is a complex mixture of hydrocarbons, whose properties strongly depend on the source and type of crude oil and refining processes. From a technical standpoint, intensive investigations carried out by the construction sector, above all by road researchers, have attempted to understand relationships between asphalt binder chemical structure, morphology and physical characteristics. Nevertheless, one challenge that the advance research on asphalt products actually face is to transfer this extremely high level of knowledge to applied industrial technologies for finding easy-to-use, quick and cost-effective test methods for quality control and identification of asphalt binders at refinery, terminal and plant. Thus, this paper focused on the development of a protocol for fingerprinting, including identification and discrimination, of asphalt cements using two different electronic noses (e-noses), also known as artificial olfactory systems (AOS). E-nose is a biomimetic non-destructive intelligent sensing instrument, which is designed to mimic the human sense of smell to detect, compare and classify odor sample, producing a qualitative output (fingerprint). Results suggested that a complementary combination of electronic nose technique and well-established analytical methodologies could be successfully used for the identification and discrimination of roofing and paving grade asphalt cements. Specifically, both sensing instruments were able to perform a good discrimination between products characterized by a different chemical nature and to verify the refinery process stability during production and a batch-to-batch crude oil consistency.

A Neuro-Inspired Spike Pattern Classifier

The emergence of post-silicon nano-devices and the coming trillion-sensor era driven by the Internet of Things have led to a search for alternative computational paradigms that can efficiently derive useful information from abundant data while enable efficient hardware implementations under significant device variations. Such computational paradigms may emerge as we explore the information processing in biological sensory systems, which achieve unparalleled performance and energy efficiency with mediocre and unreliable components. This paper presents a neuro-inspired spike pattern classifier for an artificial olfactory system, where an analog feature extraction gas sensing front end first converts sensor array information into spike patterns. The spike pattern classifier consists of a transformation of the input spikes into high-dimensional sparse vectors and a cortical memory model. The transformation is based on a random sampling scheme that can be efficiently performed with circuits exhibiting large parametric variations. An associative memory is used to perform fast and efficient storage and retrieval of the sparse vectors. The classifier achieves an output retrieval fidelity of 0.97 with a Fano factor ( $\sigma /\mu $ ) of 0.0143 when it is implemented with delay cells having a delay Fano factor of 1.6759. This is 117 times of reduction in parametric spread. It is also robust against operation failures: the output fidelity is still greater than 0.8 even when the probability of operation failure reaches 0.45.

Influence of chemical additives and wax modifiers on odor emissions of road asphalt

The use of a broad range of asphalt additives is a well-established practice in road pavement engineering for the production of high-performance hot-mix and warm-mix asphalt mixtures. The study aimed to verify and to assess, though an analytical-sensory approach based on artificial olfactory system (AOS), the effects of five different asphalt additives (chemical additives, odor suppressant agent and wax modifiers) on the odorous patterns of asphalt emissions at typical mixing and laying temperatures. The AOS has made possible to identify a specific odor fingerprint of each additive. However, once added to asphalt, these agents did not establish with binder effects of synergy, additivity or antagonism, but appear to be as neutral elements by an odorous point of view. The odorous patterns of emissions generated by heating neat asphalt at various temperatures in laboratory scale tend to coincide with those of asphalt/additive mixtures, underlining how the bituminous binder odor resulted to be hiding or masking compared to that of only-additives.

Effects of Center Metals in Porphines on Nanomechanical Gas Sensing.

Porphyrin is one of the most promising materials for realizing a practical artificial olfactory sensor system. In this study, we focus on non-substituted porphyrins—porphines—as receptor materials of nanomechanical membrane-type surface stress sensors (MSS) to investigate the effect of center metals on gas sensing. By omitting the substituents on the tetrapyrrole macrocycle of porphyrin, the peripheral interference by substituents can be avoided. Zinc, nickel, and iron were chosen for the center metals as these metalloporphines show different properties compared to free-base porphine. The present study revealed that iron insertion enhanced sensitivity to various gases, while zinc and nickel insertion led to equivalent or less sensitivity than free-base porphine. Based on the experimental results, we discuss the role of center metals for gas uptake from the view point of molecular interaction. We also report the high robustness of the iron porphine to humidity, showing the high feasibility of porphine-based nanomechanical sensor devices for practical applications in ambient conditions.

Identification of trace amounts of detergent powder in raw milk using a customized low-cost artificial olfactory system: A novel method

One of the common concerns in quality assurance of raw milk is the use of antimicrobial agents for reducing the microbial population. For this purpose, different kinds of agents may be added to raw milk like detergents. This illegal practice is harmful to human health and has ethical and serious sanitary consequences. In this study, an artificial olfactory machine (electronic nose) was developed based on eight metal oxide semiconductor sensors (MOS) and its ability to detect the presence of detergent powder in raw milk was investigated. Three features (area under the curve, relative response, and slope) were extracted from each sensor response and three baseline manipulation techniques (differential, relative and fractional) were used to correct the sensor responses. The multivariate analysis of variance (MANOVA) was employed to optimize the data matrix. MANOVA showed that the feature of “area under the curve” along with differential baseline correction method is the best combination for distinguishing different levels of the adulteration in milk. Based on the results, principal component analysis (PCA) with the first two PCs explains 91% of the variations. Linear discriminant analysis (LDA), support vector machine (SVM) and adaptive neuro-fuzzy inference system (ANFIS) method were employed for further qualitative classification. The result showed that the best performance (90%) was achieved by using the nu-SVM with Radial Basis Function (RBF) kernel function when the data collected from independent experiments were used for validation. The study demonstrated the potential of an electronic nose as a fast, effective and feasible method to detect detergent powder in raw milk.

Smart-Trace: A Future Security Aid Using Body Odor Sensing of an Individual with IoT

This future Internet era involves all the technologies-sensors, artificial intelligence, pattern recognition and machine learning to transform into a single product that can be operated from anywhere and at any time. Sensor technology widens its scope by introducing novel recognition methods for identifying an individual. Every Individual is a huge repository of unique metrics such as finger-print, Iris, finger-vein, hand geometry and much more for identifying themselves. Smell print is a state-of-art technique to perform authentication, verification and validation of a person to apply in fields of human tracking, criminal investigation and canine training etc. This paper introduces smart trace - a service to locate the human corpses as well as the live body detection technique during disasters. It achieves through E-Nose and a cogno-detective mechanism which is trained to work as artificial olfaction system that senses individual's body odor.

An Investigation into Spike-Based Neuromorphic Approaches for Artificial Olfactory Systems.

The implementation of neuromorphic methods has delivered promising results for vision and auditory sensors. These methods focus on mimicking the neuro-biological architecture to generate and process spike-based information with minimal power consumption. With increasing interest in developing low-power and robust chemical sensors, the application of neuromorphic engineering concepts for electronic noses has provided an impetus for research focusing on improving these instruments. While conventional e-noses apply computationally expensive and power-consuming data-processing strategies, neuromorphic olfactory sensors implement the biological olfaction principles found in humans and insects to simplify the handling of multivariate sensory data by generating and processing spike-based information. Over the last decade, research on neuromorphic olfaction has established the capability of these sensors to tackle problems that plague the current e-nose implementations such as drift, response time, portability, power consumption and size. This article brings together the key contributions in neuromorphic olfaction and identifies future research directions to develop near-real-time olfactory sensors that can be implemented for a range of applications such as biosecurity and environmental monitoring. Furthermore, we aim to expose the computational parallels between neuromorphic olfaction and gustation for future research focusing on the correlation of these senses.

Analytical assessment of asphalt odor patterns in hot mix asphalt production

During the various stages of asphalt paving mixtures production and placement, as a result of binder heating, a complex mixtures of hydrocarbons and volatile organic compounds (VOCs) is released into the atmosphere. As far as hot mix asphalt plants is concerned, several standards and directives were drawn up for limiting the emissions of certain specific substances, without setting any specific limit on the odorous perspective. But, odor is increasingly considered an atmospheric pollutant that can have a significant negative impact on both quality of life and economic activity. The odorous flows of HMA emitted in plants or during the paving operations and hot-in-place recycling processes in the worksites can severely limit the usability of the territory. Advances in sensor technology have made possible the development of artificial olfactory systems (AOS), which are devices designed to mimic the human olfactory system capable of characterizing osmogene mixtures. However, their potentialities have not been explored until now at any stages of asphalt and asphalt mixtures production chain. Thus, the main purpose of this study was an analytical-sensory characterization, mainly based on AOS approach, of asphalt emissions generated during the various stages of road pavement construction. The analytical and sensory analyses have firstly demonstrated the effective application of these instruments in the pavement engineering sector: an odor fingerprint of asphalt emissions, specific for each type of binder and temperature class, was determined. Thanks to the photoionization, a technique which allow the detection of organic compounds in gaseous mixtures, a pseudo-hyperbolic relationship between the release of airborne substances in asphalt emissions and the heating temperature was identified; whereas the AOS demonstrated that this increase of VOCs corresponded contextually to a change in their odorous patterns. Moreover, the existence of variations in the odor fingerprints of different asphalt binders heated at the same temperature was assessed. Through a specific statistical approach for the treatment and post-processing of data and the following elaboration of a geometric based procedure for the determination of the numerical inter-class separation, a quantitative value to the purely qualitative response of the AOS was assigned.

A Bio-Inspired Analog Gas Sensing Front End

With the Internet of Things (IoT) paradigm promising to deploy trillions of sensors, the search is on for effective means to efficiently derive useful information from the flood of sensor data through efficient hardware preprocessing. Of particular interest are computational paradigms that get their inspiration from biological sensory systems that seamlessly extract relevant information through highly efficient analog signal processing. Functions, such as feature extraction, learning, or recognition, could especially benefit from bio-inspired architectures. As an example in the case, this paper presents a bio-inspired analog gas sensing frontend for an artificial olfactory system. The analog front end implements a novel trainable feature extraction algorithm for metal-oxide gas sensor arrays. The algorithm extracts one composite feature of all analytes by performing the gradient decent algorithm during training and transforms the sensor responses into concentration-invariant spike patterns. An integrated circuit realization of the algorithm, implemented in a 65-nm CMOS technology, supports six-input channels, uses subthreshold analog circuits, and consumes 519-nW/channel in the training mode, and 463-nW/channel in the recognition mode.

Colorimetric sensor array-based artificial olfactory system for sensing Chinese green tea’s quality: A method of fabrication

ABSTRACTTea quality is often evaluated by experienced tea tasters; however, their assessments are subjective, being influenced by their individual physiological and psychological condition. Herein, we fabricated a colorimetric sensor array-based artificial olfactory system for sensing the quality of Chinese green tea. First, the colorimetric sensors array was man-made using printing 12 chemically responsive dyes (9 porphyrins, metalloporphyrins and 3 pH indicators) on silica-gel flat plate. The plate was exposed to volatile organic compounds, and the colour changes in each sample were obtained by distinguishing between the images of sensor array before and after contact with tea sample. The values of colour composition changes were extracted from the dyes’ colour sections. Multivariate calibrations were applied through principal component analysis and back propagation artificial neural network (BP-ANN) for modelling. The optimum BP-ANN model was obtained with nine principal components, and the discrimination rate was equal to 85% and 86% in the calibration and prediction sets, respectively. We thus conclude that the low cost colorimetric sensor array-based artificial olfactory technique has great potential for application in intelligent evaluation of the quality of green tea.