Space and time sensor fusion using an active camera for mobile robot navigation

We propose a sensor-fusion technique where the data sets for previous moments are properly transformed and fused into the current data sets to allow accurate measurements, such as the distance to an obstacle or the location of the service robot itself. In conventional fusion schemes, measurements are dependent on the current data sets. As a result, more sensors are required to measure a certain physical parameter or to improve the accuracy of a measurement. However, in this approach, instead of adding more sensors to the system, the temporal sequences of the data sets are stored and utilized to improve the measurements. The theoretical basis is illustrated by examples, and the effectiveness is proved through simulations. Finally, the new space and time sensor fusion (STSF) scheme is applied to the control of a mobile robot in an unstructured environment and a structured environment.This work was presented in part at the 8th International Symposium on Artificial Life and Robotics, Oita, Japan, January 24–26, 2003     

一种主动式全景立体视觉传感器设计

现有的三维激光扫描仪无法实时、全自动地计算空间三维信息,需要对获得的点云数据进行配准.为解决这些问题,研制了一种新型主动式全景立体视觉传感器( ASODVS),并且从影响ASODVS测距精度的各个因素进行了分析.实验结果表明:设计的ASODVS能自动、实时地进行空间物点深度测量,并得到相应的点云数据.     

激光视觉传感器对不连续焊缝的识别

船舶制造中,为了对小格子船舱中排水孔形成的不连续焊缝进行跟踪,在机器人上安装视觉传感器提取排水孔的起点、终点特征信号。介绍了图像的采集方法和图像处理过程。对含有焊缝信息的图片先进行去噪预处理,再进行阈值分割,求取连通区与面积;细化图像,提取出激光线条的骨架;然后再进行霍夫变换,提取特征直线;最后求得两直线的交点,即焊缝的特征点,从而识别排水孔和焊缝。在模拟船舱结构的工件上进行不连续焊缝跟踪实验,实现了不连续焊缝的识别。     

Deploying a wireless sensor network on an active volcano

Augmenting heavy and power-hungry data collection equipment with lighten smaller wireless sensor network nodes leads to faster, larger deployments. Arrays comprising dozens of wireless sensor nodes are now possible, allowing scientific studies that aren't feasible with traditional instrumentation. Designing sensor networks to support volcanic studies requires addressing the high data rates and high data fidelity these studies demand. The authors' sensor-network application for volcanic data collection relies on triggered event detection and reliable data retrieval to meet bandwidth and data-quality demands.     

A self-correcting active pixel sensor using hardware and software correction

Active pixel sensor (APS) CMOS technology reduces the cost and power consumption of digital imaging applications. We present a highly reliable system for the production of high-quality images in harsh environments. The system is based on a fault-tolerant architecture that effectively combines hardware redundancy in the APS cells and software correction techniques.     

Prototyping structural description using an inductive learning program

Character recognition systems can contribute tremendously to the advancement of the automation process and can improve the interaction between man and machine in many applications, including office automation, cheque verification and a large variety of banking, business and data entry applications. The main theme of this paper is the automatic recognition of hand‐printed Arabic characters using machine learning. Conventional methods have relied on hand‐constructed dictionaries which are tedious to construct and difficult to make tolerant to variation in writing styles. The advantages of machine learning are that it can generalize over the large degree of variation between writing styles and recognition rules can be constructed by example. The system was tested on a sample of handwritten characters from several individuals whose writing ranged from acceptable to poor in quality and the correct average recognition rate obtained using cross‐validation was 89.65%. © 2000 John Wiley & Sons, Inc.     

Presentation and description of two classification methods using data fusion based on sensor management

The application of multi-sensor fusion, which aims at recognizing a state among a set of hypotheses for object classification, is of major interest with regard to the performance improvement brought by the sensor complementarity. Nevertheless, this needs to take into account the most accurate information and take advantage of the statistical learning of the previous measurements acquired by sensors. When previous learning is not representative of real measurements provided by the sensors, the classical probabilistic fusion methods lack performance. The Dempster–Shafer theory is then introduced to face this disadvantage by integrating further information which is the context of the sensor acquisitions. In this paper, we propose a model formalism for the sensor reliability in a context that leads to two methods of integration when all the hypotheses, associated to the objects of the scene acquired by sensors, are previously learned: the first one models the integration of this further information in the fusion rule as degrees of trust and the second models the sensor reliability directly as probability mass. These two methods are based on the theory of fuzzy events. Simulations of typical cases are developed in order to define the respective validity domains of these two methods. Afterwards, we are interested in the development of these two methods in the case where the previous learning is unavailable for an object and a global method of contextual information integration can be deduced.     

Computational model for recognition of scene objects with an active sensor

In the paper, a computational model for recognition of objects in a scene image is presented. The model is based on the use of an active sensor. The structure of the object model (OM) is described. This structure is a component that stores different representations of the object and puts at user’s disposal an interface whose operations are used in the scene recognition process. Semen Yu. Sergunin. Born 1980. Graduated from the Faculty of Mathematics and Mechanics of Moscow State University in 2002. Finished postgraduate course of the Department of Computational Mathematics of the same faculty. Scientific interests include image recognition. Author of about 20 papers. Mikhail I. Kumskov. Graduated from the Faculty of Computational Mathematics and Cybernetics of Moscow State University in 1978. Received his candidate’s degree (in Physics and Mathematics) in 1981 and doctoral degree in 1997. In 1981–1997 taught at the Faculty of Computational Mathematics and Cybernetics in the special seminar on Computer Graphics and Image Processing of the Department of Automation of Systems of Computational Complexes. Since 1992 works at the laboratory of Mathematical Chemistry of the Zelinsky Institute of Organic Chemistry of the Russian Academy of Sciences. Since 1997 teaches at the Department of Computational Mathematics of the Faculty of Mathematics and Mechanics of Moscow State University. Author of more than 50 papers. Scientific interests include prediction of properties of chemical compounds, optimization of structural object representation for classification problems, and image understanding.     

Free chlorine sensing using an interferometric sensor

A waveguide interferometer based free-chlorine sensing technique has been developed. A polymer film for a specific free chlorine binding was designed, synthesized and applied on the surface of a waveguide. The material is based on cyanuric acid moieties along each repeating unit covalently tethered to poly(norbornene)s. Chlorine sensing was accomplished by measuring the refractive index change of the polymer, as a result of the reaction between cyanuric acid and free chlorine, interferometrically by the evanescent field extended above the waveguide surface. The free chlorine binding to cyanuric acid is reversible and a linear calibration curve from 0.1 to 10 mg L⁻¹ of HOCl concentration was obtained with the level of detection (LOD) and level of quantification (LOQ) of 0.047 and 0.328 mg L⁻¹ of HOCl, respectively. A free chlorine measurement with less interference from combined chlorine than DPD based colorimetric method was developed as a result of the different sensing responses of free and combined chlorine. Free chlorine residual was measured in samples collected from tap water and poultry processing waters by both optical sensor and DPD-based colorimetric method. Good agreement between both methods was observed although the levels for free chlorine measured by the optical sensor are systematically lower than the readings obtained from the DPD method. The difference might be the result of the interference from combined chlorine during the DPD measurement.     

Superresolution modeling using an omnidirectional image sensor

Recently, many virtual reality and robotics applications have been called on to create virtual environments from real scenes. A catadioptric omnidirectional image sensor composed of a convex mirror can simultaneously observe a 360-degree field of view making it useful for modeling man-made environments such as rooms, corridors, and buildings, because any landmarks around the sensor can be taken in and tracked in its large field of view. However, the angular resolution of the omnidirectional image is low because of the large field of view captured. Hence, the resolution of surface texture patterns on the three-dimensional (3-D) scene model generated is not sufficient for monitoring details. To overcome this, we propose a high resolution scene texture generation method that combines an omnidirectional image sequence using image mosaic and superresolution techniques.     

Identification using ANFIS with intelligent hybrid stable learning algorithm approaches

This paper suggests novel hybrid learning algorithm with stable learning laws for adaptive network based fuzzy inference system (ANFIS) as a system identifier and studies the stability of this algorithm. The new hybrid learning algorithm is based on particle swarm optimization (PSO) for training the antecedent part and gradient descent (GD) for training the conclusion part. Lyapunov stability theory is used to study the stability of the proposed algorithm. This paper, studies the stability of PSO as an optimizer in training the identifier, for the first time. Stable learning algorithms for the antecedent and consequent parts of fuzzy rules are proposed. Some constraints are obtained and simulation results are given to validate the results. It is shown that instability will not occur for the leaning rate and PSO factors in the presence of constraints. The learning rate can be calculated on-line and will provide an adaptive learning rate for the ANFIS structure. This new learning scheme employs adaptive learning rate that is determined by input–output data.     

SOI-CMOS compatible low-power gas sensor using sputtered and drop-coated metal-oxide active layers

In this paper, a Silicon-On-Insulator (SOI) solid-state gas-sensor with an original design of a polysilicon loop-shaped microheater fabricated on a thin-stacked dielectric membrane is presented. The microheater ensures high thermal uniformity and very low power consumption (25 mW for heating at 400°C). Sensitive films are based on tin and tungsten oxides deposited either by RF sputtering or drop coating methods. The fabricated sensors are tested to a wide variety of contaminant species and promising results are obtained. The use of completely CMOS compatible TMAH-based bulk micro-machining techniques during the fabrication process, allows easy smart gas sensor integration in SOI-CMOS technology. This makes SOI-based gas-sensing devices particularly attractive for use in handheld battery-operated gas monitors.     

Magnetic-field measurements using an integrated resonant magnetic-field sensor

The present paper introduces a magnetic-field sensor based on a resonating single-crystal silicon structure. The excitation of the resonator is achieved by the Lorentz force generated by a sinusoidal current flowing through a rectangular coil deposited on the surface of the structure. The amplitude of the vibration, which is proportional to the magnetic field, is detected by sensing capacitors. Because of the high-quality factor of the resonator, a lower detection limit of 1 nT, or even smaller might be realised when the device is vacuum-packaged. This paper describes the working principle, the fabrication procedure as well as open- and closed-loop measurement results.     

Soft authentication using an infrared ceiling sensor network

Person identification is needed to provide various personalized services at home or at the office. We propose a system for tracking persons identified at the entrance to a room in order to realize “soft authentication.” Our system can be constructed at low cost and works anytime and anywhere in a room. Through experiments, we confirmed that the system could track up to 5 persons with a high probability of correct identification, though precise identification is difficult.     

Reconstructing social interactions using an unreliable wireless sensor network

In the very active field of complex networks, research advances have largely been stimulated by the availability of empirical data and the increase in computational power needed for their analysis. These works have led to the identification of similarities in the structures of such networks arising in very different fields, and to the development of a body of knowledge, tools and methods for their study.While many interesting questions remain open on the subject of static networks, challenging issues arise from the study of dynamic networks. In particular, the measurement, analysis and modeling of social interactions are first class concerns.In this article, we address the challenges of capturing physical proximity and social interaction by means of a wireless network. In particular, as a concrete case study, we exhibit the deployment of a wireless sensor network applied to the measurement of health care workers’ exposure to tuberculosis-infected patients in a service unit of the Bichat-Claude Bernard hospital in Paris, France. This network has continuously monitored the presence of all HCWs in all rooms of the service during a three month period.We both describe the measurement system that was deployed and some early analysis on the measured data. We highlight the bias introduced by the measurement system reliability and provide a reconstruction method which not only leads to a significantly more coherent and realistic dataset but also evidences phenomena a priori hidden in the raw data. By this analysis, we suggest that a processing step is required prior to any adequate exploitation of data gathered thanks to a non-fully reliable measurement architecture.     

A tool breakage detection system using an accelerometer sensor

An on-line and in-process based monitoring system to detect tool breakage via an accelerometer was developed and successfully evaluated in an end milling operation. Prior to testing and evaluation of the tool breakage condition, a simulation model was developed. Transfer of the on-line vibration signal to the frequency domain employed the fast fourier transfer function, and set thresholds were used to determine the tool condition after various experimental tests. In comparison to other in-process methods, such as those that employ dynamometers and acoustic emission sensors, the proposed system is easy to set up and does not require changing of the mechanism. Additional benefits of the system are its high reliability and low cost. Thus, the new monitoring system is potentially useful for untended milling operations in on-line and real-time tool breakage detection in linked-cell manufacturing systems (L-CMS).     

Development of an ozone gas sensor using single-walled carbon nanotubes

This study deals with the fabrication of an ozone gas sensor using single-walled carbon nanotubes (SWCNTs) as sensing material. The SWCNTs are dispersed by N,N-dimethylformamide (DMF). The CNT-DMF solution was dropped between interdigitated electrodes’ fingers to fabricate ozone gas sensor. For ozone environment, a commercial ozone generator was introduced. To improve sensor response, the deposited carbon nanotubes network was thermally treated at high temperature in a furnace. The sensor exhibits high sensitivity to ozone gas at concentration as low as 50 ppb, and fast response time, which is promising for future commercialization of carbon nanotubes based ozone gas sensor.     

Implementing temporal integrity constraints using an active DBMS

The paper proposes a general architecture for implementing temporal integrity constraints by compiling them into a set of active DBMS rules. The modularity of the design allows easy adaptation to different environments. Both differences in the specification languages and in the target rule systems can be easily accommodated. The advantages of this architecture are demonstrated on a particular temporal constraint compiler. This compiler allows automatic translation of integrity constraints formulated in Past Temporal Logic into rules of an active DBMS (in the current version of the compiler two active DBMS are supported: Starburst and INGRES). During the compilation the set of constraints is checked for the safe evaluation property. The result is a set of SQL statements that includes all the necessary rules needed for enforcing the original constraints. The rules are optimized to reduce the space overhead introduced by the integrity checking mechanism. There is no need for an additional runtime constraint monitor. When the rules are activated, all updates to the database that violate any of the constraints are automatically rejected (i.e., the corresponding transaction is aborted). In addition to straightforward implementation, this approach offers a clean separation of application programs and the integrity checking code