Sensor Fault Tolerance Method by Using a Bayesian Network for Robot Behavior

This paper presents FTBN, a new framework that performs learning autonomous mobile robot behavior and fault tolerance simultaneously. For learning behavior in the presence of a robot sensor fault this framework uses a Bayesian network. In the proposed framework, sensor data are used to detect a faulty sensor. Fault isolation is accomplished by changing the Bayesian network structure using interpreted evidence from robot sensors. Experiments including both simulation and a real robot are performed for door-crossing behavior using prior knowledge and sensor data at several maps. This paper explains the learning behavior, optimal tracking, exprimental setup and structure of the proposed framework. The robot uses laser and sonar sensors for door-crossing behavior, such that each sensor can be corrupted during the behavior. Experimental results show FTBN leads to robust behavior in the presence of a sensor fault as well as performing better compared to the conventional Bayesian method.     

WSNs deployment framework based on the theory of belief functions

Deployment is a fundamental issue in Wireless Sensor Networks (WSNs). Indeed, the number and locations of sensors determine the topology of the WSN, which will further influence its performance. Usually, the sensor locations are precomputed based on a “perfect” sensor coverage model. However, in reality, there is an inherent uncertainty and imprecision associated with sensor readings. This issue impinges upon the success of any WSN deployment, and it is therefore important to consider it at the design stage. In contrast to existing work, this paper investigates the belief functions theory to design a unified approach for robust uncertainty-aware WSNs deployment. Specifically, we address the issue of handling uncertainty and information fusion for an efficient WSNs deployment by exploiting the belief functions reasoning framework. We present a flexible framework for target/event detection within the transferable belief model. Using this framework, we propose uncertainty-aware deployment algorithms that are able to determine the minimum number of sensors as well as their locations in such a way that full area coverage is provided. Related issues, such as connectivity, preferential coverage, challenging environments and sensor reliability, are also discussed. Simulation results, based on both synthetic data set and data traces collected in a real deployment for vehicle detection, are provided to demonstrate the ability of our approach to achieve an efficient WSNs deployment by exploiting a collaborative target/event detection scheme. Using the devised approach, we successfully deploy an experimental testbed for motion detection. The obtained results are reported, supported by comparison with other works.     

Soft sensor based on stacked auto-encoder deep neural network for air preheater rotor deformation prediction

Soft sensors have been widely used in industrial processes over the past two decades because they use easy-to-measure process variables to predict difficult-to-measure ones. Some success has been achieved by the dominant traditional methods of modeling soft sensors based on statistics, such as principal components analysis (PCA) and partial least square (PLS), but such sensors usually become inaccurate and inefficient when processing strong nonlinear data. In this paper, a new soft sensor modeling approach is proposed based on a deep learning network. First, stacked auto-encoders (SAEs) are employed to extract high-level feature representations of the input data. In the process of training each layer of a SAE, the Limited-memory Broyden-Fletcher-Goldfarb-Shanno algorithm (L-BFGS) is adopted to optimize the weights parameters. Then, a support vector regression (SVR) is added to predict the target value on the basis of the features obtained from the SAE. To improve the model performance, Genetic Algorithm (GA) is used to obtain the optimal parameters of the SVR. To evaluate the proposed method, a soft sensor model for estimating the rotor deformation of air preheaters in a thermal power plant boiler is studied. The experimental results demonstrate that the soft sensor based on the SAE-SVR algorithm is more effective than the existing methods are.     

输入不确定BELM在发酵过程软测量中的应用

为复杂的发酵过程建立软测量模型要求模型最好能够给出预测值的置信区间,以便技术人员对发酵过程的真实状况和模型的可靠性进行评估。贝叶斯极限学习机能够在实现预测的同时一并给出预测值的置信区间,因此将其用于发酵过程的软测量建模。然而,实际发酵过程中的输入数据往往带有噪声,贝叶斯极限学习机仅能处理输出含噪声的情况。针对这个问题,提出了输入不确定贝叶斯极限学习机。在原有的贝叶斯推理过程中引入输入不确定性,得到了综合考虑输入输出噪声的模型参数和预测置信区间。最后利用青霉素发酵过程进行仿真验证,建立了产物质量浓度的软测量模型,结果表明该方法预测精度高,得到的预测置信区间包含了所有真实值。     

Online monitoring and control of a cyber-physical manufacturing process under uncertainty

Recent technological advancements in computing, sensing and communication have led to the development of cyber-physical manufacturing processes, where a computing subsystem monitors the manufacturing process performance in real-time by analyzing sensor data and implements the necessary control to improve the product quality. This paper develops a predictive control framework where control actions are implemented after predicting the state of the manufacturing process or product quality at a future time using process models. In a cyber-physical manufacturing process, the product quality predictions may be affected by uncertainty sources from the computing subsystem (resource and communication uncertainty), manufacturing process (input uncertainty, process variability and modeling errors), and sensors (measurement uncertainty). In addition, due to the continuous interactions between the computing subsystem and the manufacturing process, these uncertainty sources may aggregate and compound over time. In some cases, some process parameters needed for model predictions may not be precisely known and may need to be derived from real time sensor data. This paper develops a dynamic Bayesian network approach, which enables the aggregation of multiple uncertainty sources, parameter estimation and robust prediction for online control. As the number of process parameters increase, their estimation using sensor data in real-time can be computationally expensive. To facilitate real-time analysis, variance-based global sensitivity analysis is used for dimension reduction. The proposed methodology of online monitoring and control under uncertainty, and dimension reduction, are illustrated for a cyber-physical turning process.

     

一种优化的贝叶斯估计多传感器数据融合方法

由于来自多个传感器的测量数据总是有一定程度的不确定性和不一致性,采用多传感器数据融合算法将多个节点的测量数据进行数据融合,利用数据的冗余度来减小这种不确定性,得到高可靠性的数据信息。提出了一种优化的贝叶斯估计多传感器数据融合方法,将贝叶斯估计和卡尔曼滤波器结合起来,应用于无线传感网络数据融合中。根据滤波器应用到传感数据、融合数据或者两者的方式,提出3种不同的技术,即:前向滤波法、后向滤波法和前后向滤波法。通过一个实例研究估计移动机器人的位置,验证算法的有效性。实验表明,在集中式和分布式两个方面数据融合体系结构,结合卡尔曼滤波器的贝叶斯融合算法能够有效地解决数据的不确定性和不一致性。     

Comparison of distribution strategies in uncertainty-aware catchment delineation

Delineation of drainage basins from a digital elevation model (DEM) has become a standard operation in a number of terrain analysis software packages, but limitations of the conventionally used techniques have become apparent. Firstly, the delineation methods make assumption of error-free data, which is an unreachable utopia even with modern sensor technology. Secondly, even though the computing capacity has increased dramatically during the last decades, sizes of geospatial data sets have increased simultaneously. Thus far, the typical problems arising when using uncertainty-aware geospatial analysis are 1) the computational complexity of the analysis and 2) memory allocation problems when large datasets are used. In this paper, we raise the question about the general need for developing scalable and uncertainty-aware algorithms for terrain analysis and propose improvements to the existing drainage basin calculation methods. The distributed uncertainty-aware catchment delineation methods with and without spatial partitioning of the DEM are introduced and the performance of the methods in different cases are compared.     

Environmental map building for a mobile robot using infrared range-finder sensors

This paper presents a methodology for building a high-accuracy environmental map using a mobile robot. The design approach uses low-cost infrared range-finder sensors incorporating with neural networks. To enhance the map quality, the errors occurring from the sensors are corrected. The non-linearity error of the sensors is compensated using a backpropagation neural network and the random error of readings including the uncertainty of the environment is taken into a sensor model as a probabilistic approach. The map is represented by an occupancy grid framework and updated by the Bayesian estimation mechanism. The effectiveness of the proposed method is verified through a series of experiments.     

Data-driven soft sensor development based on deep learning technique

In industrial process control, some product qualities and key variables are always difficult to measure online due to technical or economic limitations. As an effective solution, data-driven soft sensors provide stable and reliable online estimation of these variables based on historical measurements of easy-to-measure process variables. Deep learning, as a novel training strategy for deep neural networks, has recently become a popular data-driven approach in the area of machine learning. In the present study, the deep learning technique is employed to build soft sensors and applied to an industrial case to estimate the heavy diesel 95% cut point of a crude distillation unit (CDU). The comparison of modeling results demonstrates that the deep learning technique is especially suitable for soft sensor modeling because of the following advantages over traditional methods. First, with a complex multi-layer structure, the deep neural network is able to contain richer information and yield improved representation ability compared with traditional data-driven models. Second, deep neural networks are established as latent variable models that help to describe highly correlated process variables. Third, the deep learning is semi-supervised so that all available process data can be utilized. Fourth, the deep learning technique is particularly efficient dealing with massive data in practice.     

Deep learning-based construction equipment operators’ mental fatigue classification using wearable EEG sensor data

Operator attention failure due to mental fatigue during extended equipment operations is a common cause of equipment-related accidents that result in catastrophic injuries and fatalities. As a result, tracking operators' mental fatigue is critical to reducing equipment-related accidents on construction sites. Previously, several strategies aimed at recognizing mental fatigue with adequate accuracy, such as machine learning utilizing EEG-based wearable sensing systems, have been proposed. However, the ability to track operators’ mental fatigue for its implementation on an actual construction site is still an issue. For instance, the mobility and systemic instability of EEG sensors necessitate their application in laboratory settings rather than on actual construction sites. Furthermore, while the machine learning classifiers achieved acceptable accuracy, their input is limited to manually developed EEG features, which may compromise the models’ performance on real construction sites. Accordingly, the current research proposes the viability of a construction site strategy that uses flexible headband-based sensors for acquiring raw EEG data and deep learning networks to recognize operators' mental fatigue. To serve this purpose, a one-hour excavator operation by fifteen operators was conducted on a construction site. The NASA-TLX score was used as the ground truth of mental fatigue, and brain activity patterns were recorded using a wearable EEG sensor. The raw EEG data was then used to develop deep learning-based classification models. Finally, the performance of deep learning models, i.e., long short-term memory, bidirectional LSTM, and one-dimensional convolutional networks, was investigated using accuracy, precision, recall, specificity, and an F1-score. The findings indicate that the Bi-LSTM model outperforms the other deep learning models with a high accuracy of 99.941% and F1-score between 99.917% and 99.993%. These findings demonstrate the feasibility of applying the Bi-LSTM model and contribute to wearable sensor-based mental fatigue recognition and classification, thus enhancing on-site health and safety operations.     

基于块稀疏贝叶斯学习的体域网心电压缩采样

为有效提高体域网的实时性和降低体域网的功耗,提出一种基于块稀疏贝叶斯学习的体域网心电压缩采样方法。该方法在体域网框架下,利用压缩采样理论,在体域网的传感节点利用二进制随机观测矩阵对心电信号进行压缩采样,远程监护中心获得采样值之后,利用块稀疏贝叶斯学习重构算法和离散余弦稀疏变换矩阵对心电信号进行重构。实验结果表明,当心电信号压缩率在70%~90%时,基于块稀疏贝叶斯学习的重构算法要比其他重构算法的重构信噪比高出3 dB~21 dB。该方法能有效减少数据采样,减轻后续的数据存储、数据传输压力,提高体域网的实时性。同时该方法具有功耗低,易于硬件实现的优点。     

Streaming parallel variational Bayesian supervised factor analysis for adaptive soft sensor modeling with big process data

Time-varying and state shifting are two of the main process factors that cause poor prediction performance of soft sensors. Adaptive soft sensor is commonly an alternative practice to ensure high predictive accuracy. However, the large scale of process data often leads to inefficiency of model updating. In this paper, a streaming variational Bayesian supervised factor analysis (S-VBSFA) model is first proposed to capture the process time-varying and state shifting features through online updating of the posterior of model parameters. During the updating process, the symmetric Kullback–Leibler (SKL) divergence is utilized to determine priors of the next variation Bayesian inference. To improve the modeling efficiency for large-scale process data, the parallel computing strategy is further applied to the streaming model. As a result, the proposed streaming parallel VBSFA (SP-VBSFA) algorithm not only relieves the computing pressure of modeling big process data, but also improves the prediction accuracy and further reduces the tracking time delay for process variations. Two case studies demonstrate the superiority of the proposed method, compared to conventional methods.     

Decision making and uncertainty management in a 3D reconstruction system

This paper presents a control structure for a general-purpose image understanding system. It addresses the high level of uncertainty in local hypotheses and the computational complexity of image interpretation. The control of vision algorithms is done by an independent subsystem that uses Bayesian networks and utility theory to compute marginal value of information and selects the algorithm with the highest value of information. It is shown that the knowledge base can be acquired using learning techniques and the value-driven approach to the selection of vision algorithms leads to performance gains.     

A frequentist approach to mapping under uncertainty

An asynchronous stochastic approximation based (frequentist) approach is proposed for mapping using noisy mobile sensors under two different scenarios: (1) perfectly known sensor locations and (2) uncertain sensor locations. The frequentist methodology has linear complexity in the map components, is immune to the data association problem and is provably consistent. The frequentist methodology, in conjunction with a Bayesian estimator, is applied to the Simultaneous Localization and Mapping (SLAM) problem of Robotics. Several large maps are estimated using the hybrid Bayesian/Frequentist scheme and results show that the technique is robust to the computational and performance issues inherent in the purely Bayesian approaches to the problem.     

在线场景更新的多阶段非线性模型预测聚合反应控制

针对聚合过程中时不变不确定性参数不能直接估计的情况,导致的多阶段非线性模型预测控制中场景树生成的合理性问题,提出一种基于贝叶斯概率加权的在线场景更新算法.该方法利用前一时间步中每个场景的模型预测信息和过程状态测量信息计算对应场景的概率权重,然后通过合适的自适应步长在线更新场景树中不确定性的离散实现场景.所提方法在保证过程约束满足的同时,逐渐缩小不确定性集合逼近不确定性的真实值,从而降低保守性,提升控制器性能.通过多个批次的半间歇聚合反应过程实例仿真结果表明,所提出的方法可以有效降低批次反应时间,提高生产效率.     

Using artificial intelligence to predict surface roughness in deep drilling of steel components

A predictive model is presented to optimize deep drilling operations under high speed conditions for the manufacture of steel components such as moulds and dies. The input data include cutting parameters and axial cutting forces measured by sensors on the milling centres where the tests are performed. The novelty of the paper lies in the use of Bayesian Networks that consider the cooling system as an input variable for the optimization of roughness quality in deep drilling operations. Two different coolant strategies are tested: traditional working fluid and MQL (Minimum Quantity Lubrication). The model is based on a machine learning classification method known as Bayesian networks. Various measures used to assess the model demonstrate its suitability to control this type of industrial task. Its ease of interpretation is a further advantage in comparison with other artificial intelligence tools, which makes it a user-friendly application for machine operators.     

Data supplement for a soft sensor using a new generative model based on a variational autoencoder and Wasserstein GAN

In industrial process control, measuring some variables is difficult for environmental or cost reasons. This necessitates employing a soft sensor to predict these variables by using the collected data from easily measured variables. The prediction accuracy and computational speed in the modeling procedure of soft sensors could be improved with adequate training samples. However, the rough environment of some industrial fields makes it difficult to acquire enough samples for soft sensor modeling. Generative adversarial networks (GANs) and the variational autoencoder (VAE) are two prominent methods that have been employed for learning generative models. In the current work, the VA-WGAN combining VAE with Wasserstein generative adversarial networks (WGAN) as a generative model is established to produce new samples for soft sensors by using the decoder of VAE as the generator in WGAN. An actual industrial soft sensor with insufficient data is used to verify the data generation capability of the proposed model. According to the experimental results, the samples obtained with the proposed model more closely resemble the true samples compared with the other four common generative models. Moreover, the insufficiency of the training data and the prediction precision of soft sensors could be improved via these constructed samples.     

Development and industrial application of soft sensors with on-line Bayesian model updating strategy

This paper deals with the issues associated with the development of data-driven models as well as model update strategy for soft sensor applications. A practical yet effective solution is proposed. Key process variables that are difficult to measure are commonly encountered in practice due to limitations of measurement techniques. Even with appropriate instruments, some measurements are only available through off-line laboratory analysis with typical sampling intervals of several hours. Soft sensors are inferential models that can provide continuous on-line prediction of hidden variables; such models are capable of combining real-time measurements with off-line lab data. Due to the prevalence of plant-model mismatch, it is important to update the model using the latest reference data. In this paper, parameters of data-driven models are estimated using particle filters under the framework of expectation–maximization (EM) algorithms. A Bayesian methodology for model calibration strategy is formulated. The proposed framework for soft sensor development is applied to an industrial process to provide on-line prediction of a quality variable.     

On the use of Bayesian Networks to develop behaviours for mobile robots

Bayesian Networks are models which capture uncertainties in terms of probabilities that can be used to perform reasoning under uncertainty. This paper presents an attempt to use Bayesian Networks as a learning technique to manage task execution in mobile robotics. To learn the Bayesian Network structure from data, the K2 structural learning algorithm is used, combined with three different net evaluation metrics. The experiment led to a new hybrid multiclassifying system resulting from the combination of 1-NN with the Bayesian Network, that allows one to use the power of the Bayesian Network while avoiding the computational burden of the reasoning mechanism — the so-called evidence propagation process. As an application example we present an approach of the presented paradigm to implement a door-crossing behaviour in a mobile robot using only sonar readings, in an environment with smooth walls and doors. Both the performance of the learning mechanism and the experiments run in the real robot-environment system show that Bayesian Networks are valuable learning mechanisms, able to deal with the uncertainty and variability inherent to such systems.     

Variable activation function extreme learning machine based on residual prediction compensation

For solving the problem that extreme learning machine (ELM) algorithm uses fixed activation function and cannot be residual compensation, a new learning algorithm called variable activation function extreme learning machine based on residual prediction compensation is proposed. In the learning process, the proposed method adjusts the steep degree, position and mapping scope simultaneously. To enhance the nonlinear mapping capability of ELM, particle swarm optimization algorithm is used to optimize variable parameters according to root-mean square error for the prediction accuracy of the mode. For further improving the predictive accuracy, the auto-regressive moving average model is used to model the residual errors between actual value and predicting value of variable activation function extreme learning machine (V-ELM). The prediction of residual errors is used to rectify the prediction value of V-ELM. Simulation results verified the effectiveness and feasibility of this method by using Pole, Auto-Mpg, Housing, Diabetes, Triazines and Stock benchmark datasets. Also, it was implemented to develop a soft sensor model for the gasoline dry point in delayed coking and some satisfied results were obtained.