Multi-robot cooperation and data fusion in map-building

In order to solve the problem that it is difficult to achieve better multi-robot coordination operation using the original control system in multi-robot cooperative systems, this paper conducts system modeling, control framework design and software development and verification experiments for multi-robot cooperative systems. We firstly model the system composed of multiple robots and operated objects. The coordination between any two robots can be divided into two categories: loose coordination and tight coordination according to the constraint relationship between the robots and the objects, based on which the kinematics and dynamics analysis of the multi-robot system is carried out. A multi-robot cooperative control framework is designed, which contains modules for task planning, force and position assignment, sensor information acquisition, position/force control and synchronization clock. On the basis of the multi-robot cooperative control system, the robot cooperative control software is designed and developed, and the software system contains, with excellent characteristics. Finally, an experimental environment containing three large-load EFORT industrial robots was established and experiments were conducted using the developed multi-robot cooperative control software to verify the usability and reliability of the software system.

To build an early fire alarm system, an important approach to detect whether a fire occurs is monitoring the fire characteristics parameters. Multi-sensor data fusion is the key technology to process the data collected by the sensors. In this paper, we present a data fusion method to process the data collected by distributed fire detectors. The method is based on rough sets and evidence theory. We explain the method and use an example to demonstrate its decision process under the Chinese standard on raked fire SH4 and smoldering fire SH1. The results prove that the method is fault-tolerant and stable.

Theory of evidence is a mathematical theory which allows one to reason with uncertainty and which suggests a way for combining uncertain data. This is the reason why it is used as a basic tool for multisensor data fusion in the situation assessment process. Although numerical people know this formalism quite well and its use in multisensor fusion, it is not the case for people used to manipulating logical formalisms. The article intends to give them the key for understanding Theory of Evidence and its use in multisensor data fusion. It does this by giving a logical interpretation of this formalism when the numbers are rational, and also by reformulating, in a particular case, one model defined by A. Appriou (1997) for multisensor data fusion.

This study proposed a novel way to calculate Basic Probability Assignment(BPA), which is crucial in the data fusion. The study mapped sensing signal with a linear discriminant function analysis and assessed the sizes with time. This was beneficial to get a clue for context inference with using the Dempster-Shafer Theory and to determine BPA based on the size changing of mapping data in time intervals. The study provides with the way of context inference for fast detecting a local event that affects the whole area. Data fusion process with multi-sensors has caught increasingly more attention nowadays. Fuzzy Theory (FT) and Dempster-Shafer Theory (DST) have been used much as a method to process heterogeneous signals and data fusion although these theories were designed originally to express the ambiguity and uncertainty of events occurring in the real world. Multi-sensors data fusion process is applied to acquire information with high quality by using some kinds of sensors in biometrics, geographic information, network security, robot control, etc. Multi-sensors data fusion is also important to acquire context information with high quality in the Ubiquitous Sensor Network(USN) data process. Multi-sensors data fusion for acquiring context information in the sensor network has become big data process in the data streaming conditions. The studies in the data streaming conditions first consider searching, classifying, and clustering the characteristics of continuous data that are not saved in a server. However, it is necessary to infer the changing context in real time through the data patterns acquired from the real time data. For example, we need to infer situations urgently and efficiently without holding to analyze when a blaze in a large area occurs, objects move around the road, or some risk factors threat the security in our residential environment. To do so, it is necessary to infer the context based on the rapid analysis of data patterns by time intervals in the data stream. We could obtain a clue of the simple and early context inference from the changing patterns with time of the sensing data detected and reported by sensors. The study presents a way to infer context information by discriminating event information reported to a host through sink nodes in the data streaming environment. It is also suggested that the results from discriminating event information are applied to BPA determination with using DST. It is reasonable to determine BPA based on the detection of patterns appeared in

Traditional Intrusion Detection System (IDS) focus on low-level attacks or anomalies, and too many alerts are produced in practical application. Based on the D-S Evidence Theory and its data fusion technology, a novel detection data fusion model-IDSDFM is presented. By correlating and merging alerts of different types of IDSs, a set of alerts can be partitioned into different alert tracks such that the alerts in the same alert track may correspond to the same attack. On the base of it, the current security situation of network is estimated by applying the D-S Evidence Theory, and some IDSs in the network are dynamically adjusted to strengthen the detection of the data which relate to the attack attempts. Consequently, the false positive rate and the false negative rate are effectively reduced, and the detection efficiency of IDS is improved.

This work introduces the use of data fusion in the field of DDoS anomaly detection. We present Dempster-Shafer Theory of Evidence (D-S), the mathematical foundation for the development of a novel DDoS detection engine. Based on a data fusion paradigm, we combine evidence generated from multiple simple metrics to feed our D-S inference engine and detect attacks on a single network element (high bandwidth link).The main advantages of our approach are the modeling power of the Theory of Evidence in expressing beliefs in some hypotheses, its flexibility to handle uncertainty and ignorance and its ability to provide quantitative measurement of the belief and plausibility in our detection results. Furthermore we propose a system that can be trained (supervised learning) with minimum human effort, using in parallel expert knowledge about each metric detection ability.We evaluate our detection engine prototype through an extensive set of experiments on an operational network using real network traffic, with the use of a popular DDoS attack generator. Based on these results we discuss the performance of our D-S scheme in contrast to simple thresholds on single metrics, as well as against an alternative data fusion technique based on an Artificial Neural Network. We conclude that our data fusion is a promising approach that significantly increases the DDOS detection rate (true positives) while keeping the false positive alarm rate low.

In order to improve the robustness of the multi-sensor data fusion system, a novel data fusion algorithm is proposed based on Dempster-Shafer theory. Firstly the accuracy of the measured data are obtained according to relative distance between data, then the appropriate weight assignment method is proposed inspired by basic probability assignment, thirdly evidence combination rule is applied to adjust the weight of the data obtained before, finally the fusion result is obtained by weighted summation. Good performance of the proposed algorithm is demonstrated by numerical examples.

In this paper, we presented an equipment fault diagnosis method based on multi-sensor data fusion, in order to solve the problems such as uncertainty, imprecision and low reliability caused by using a single sensor to diagnose the equipment faults. We used a variety of sensors to collect the data for diagnosed objects and fused the data by using D-S evidence theory, according to the change of confidence and uncertainty, diagnosed whether the faults happened. Experimental results show that, the D-S evidence theory algorithm can reduce the uncertainty of the results of fault diagnosis, improved diagnostic accuracy and reliability, and compared with the fault diagnosis using a single sensor, this method has a better effect.

Interpreting drinking water quality in the distribution system using Dempster–Shafer theory of evidence

Although a huge amount of information is available in all kinds of marine safety databases, only a small part can be used directly. The disorganized information from different sources leads to a mixture of format and definition. In the work reported in this paper, the Dempster–Shafer theory (DST) of evidence was applied to combine evidence (i.e., a piece of information that supports a claim) from different sources. The method is regarded as a generalization of the Bayesian theory and can avoid two difficulties in classical probability theory: handling the conflicting information and assigning prior probabilities. The work of data fusion was demonstrated first by a decision fusion problem that involved the reconciliation of contradictory expert reports. The DST can provide a decision maker with a comprehensive result through the combination of different experts’ opinions. Second, fusion was conducted of two representative maritime incident databases: that of the Global Integrated Shipping Information System and that of the International Chamber of Commerce. Although the records in the databases had some defects (e.g., disorder, error, contradiction), the DST was able to work effectively and calculate an uncertainty interval of incident.

This article presents a novel assessment framework to identify the health condition of wood utility poles. The innovative approach is based on the integration of data mining and machine learning methods and combines advanced signal processing, multi-sensor data fusion and decision ensembles to classify different damage condition types of wood poles. In the proposed framework, wavelet packet analysis is employed to transform captured multi-channel stress wave signals into energy information, which is consequently compressed by principal component analysis to extract a feature vector. Furthermore, support vector machine multi-classifier, optimized by genetic algorithm, is designed to identify the pole condition type. Finally, evidence theory is applied to fuse different assessment results from different sensors for a final decision. For validation of the proposed approach, the wood pole specimens with three common damage condition types are tested using a novel multi-sensor narrow-band frequency-excitation non-destructive testing system in the laboratory. The final experimental analysis results confirm that the proposed approach is capable of making full use of multi-sensor information and providing an effective and accurate identification on types of conditions in wood poles.

Multi-resource fusion of nondestructive evaluation data for bridge deck assessment using discrete wavelet transform and Dempster-Shafer theory

The theory of belief functions has been extensively utilized in many practical applications involving decision making. One such application is the classification of target based on the pieces of information extracted from the individual attributes describing the target. Each piece of information is usually modeled as the basic probability assignment (BPA), also known as the mass function. The determination of the BPA has remained an open problem. Although fuzzy membership functions such as triangular and Gaussian functions have been widely used to model the likelihood estimation function based on the historical data, it has been observed that less emphasis has been placed on the impact of the spread of the membership function on the decision accuracy of the reasoning process. Conflict in the combination of BPAs may arise due to poor characterization of fuzzy membership functions to induce belief mass. In this work, we propose a multisensor data fusion within the framework of belief theory for target classification where shape/spread of the membership function is adjusted during the training/modeling stage to improve on the classification accuracy while removing the need for the computation of the credibility. To further enhance the performance of the proposed method, the reliability factor is deployed not only to effectively manage the possible conflict among participating bodies of evidence for better decision accuracy but also to reduce the number of sources for improved efficiency. The effectiveness of the proposed method was evaluated using both the real-world and the artificial datasets.

Within the framework of pattern recognition, many methods of classification were developed. More recently, techniques using the Dempster-Shafer's theory or evidence theory tried to deal with the problem related to the management of the uncertainty and data fusion. We propose a classification method based on Dempster-Shafer's theory and information criteria. After an original basic belief assignment method, we introduce an attenuation factor of belief functions based on the dissimilarity between probability distributions. Results on synthetic and real world data sets are given in order to illustrate the proposed methodology.

Data association is an important task in Multiple Target Tracking (MTT) systems. The purpose is to assign the sensor detections to the known objects. However, sensor data can be inaccurate and incomplete. The Evidential theory provides interesting tools to manage ignorance and data imperfection. In fact, the Evidential theory is considered as a generalized form of the Bayesian theory which enables reasoning with uncertainty. The Evidential data association quantifies sensor information by belief masses. Subsequently, the combination of these masses provides more accurate data to make-decision on associations. Most published studies focus on decision-making algorithms. In this paper, the importance is done to belief masses estimation which is a crucial step in data fusion. There are two estimation models: antagonist and non-antagonist. The objective of this paper is to benchmark both models on real-data. The obtained results show that the non-antagonist mass function is more suited for data association problems.