Search results

This study aims to develop a data fusion method for powder-bed fusion (PBF) process monitoring based on process image information. The data fusion method can help improve process condition identification performance, which can provide guidance for further PBF process monitoring and control system development.

Design of reliable process monitoring systems is an essential approach to solve PBF built quality. A data fusion framework based on support vector machine (SVM), convolutional neural network (CNN) and Dempster-Shafer (D-S) evidence theory are proposed in the study. The process images which include the information of melt pool, plume and spatters were acquired by a high-speed camera. The features were extracted based on an appropriate image processing method. The three feature vectors corresponding to the three objects, respectively, were used as the inputs of SVM classifiers for process condition identification. Moreover, raw images were also used as the input of a CNN classifier for process condition identification. Then, the information fusion of the three SVM classifiers and the CNN classifier by an improved D-S evidence theory was studied.

The results demonstrate that the sensitivity of information sources is different for different condition identification. The feature fusion based on D-S evidence theory can improve the classification performance, with feature fusion and classifier fusion, the accuracy of condition identification is improved more than 20%.

An improved D-S evidence theory is proposed for PBF process data fusion monitoring, which is promising for the development of reliable PBF process monitoring systems.

Intent detection (ID) and slot filling (SF) are two important tasks in natural language understanding. ID is to identify the main intent of a paragraph of text. The goal of SF is to extract the information that is important to the intent from the input sentence. However, most of the existing methods use sentence-level intention recognition, which has the risk of error propagation, and the relationship between intention recognition and SF is not explicitly modeled. Aiming at this problem, this paper proposes a collaborative model of ID and SF for intelligent spoken language understanding called ID-SF-Fusion.

ID-SF-Fusion uses Bidirectional Encoder Representation from Transformers (BERT) and Bidirectional Long Short-Term Memory (BiLSTM) to extract effective word embedding and context vectors containing the whole sentence information respectively. Fusion layer is used to provide intent–slot fusion information for SF task. In this way, the relationship between ID and SF task is fully explicitly modeled. This layer takes the result of ID and slot context vectors as input to obtain the fusion information which contains both ID result and slot information. Meanwhile, to further reduce error propagation, we use word-level ID for the ID-SF-Fusion model. Finally, two tasks of ID and SF are realized by joint optimization training.

We conducted experiments on two public datasets, Airline Travel Information Systems (ATIS) and Snips. The results show that the Intent ACC score and Slot F1 score of ID-SF-Fusion on ATIS and Snips are 98.0 per cent and 95.8 per cent, respectively, and the two indicators on Snips dataset are 98.6 per cent and 96.7 per cent, respectively. These models are superior to slot-gated, SF-ID NetWork, stack-Prop and other models. In addition, ablation experiments were performed to further analyze and discuss the proposed model.

This paper uses word-level intent recognition and introduces intent information into the SF process, which is a significant improvement on both data sets.

The historical data usually consist of overlapping reports, and these reports may contain inconsistent data, which may return incorrect results of a query search. Moreover, users who issue the query may not learn of this inconsistency even after a data cleaning process (e.g. schema matching or data screening). The inconsistency can exist in different types of data, such as temporal or spatial data. Therefore, this paper aims to introduce an information fusion method that can detect data inconsistency in the early stages of data fusion.

This paper introduces an information fusion method for multi-robot operations, for which fusion is conducted continuously. When the environment is explored by multiple robots, the robot logs can provide more information about the number and coordination of targets or victims. The information fusion method proposed in this paper generates an underdetermined linear system of overlapping spatial reports and estimates the case values. Then, the least squares method is used for the underdetermined linear system. By using these two methods, the conflicts between reports can be detected and the values of the intervals at specific times or locations can be estimated.

The proposed information fusion method was tested for inconsistency detection and target projection of spatial fusion in sensor networks. The proposed approach examined the values of sensor data from simulation that robots perform search tasks. This system can be expanded to data warehouses with heterogeneous data sources to achieve completeness, robustness and conciseness.

Little research has been devoted to the linear systems for information fusion of tasks of mobile robots. The proposed information fusion method minimizes the cost of time and comparison for data fusion and also minimizes the probability of errors from incorrect results.

Welding sensor technology is the key technology in welding process, but a single sensor cannot acquire adequate information to describe welding status. This paper addresses arc sensor and sound sensor to acquire the voltage and sound information of pulsed gas tungsten arc welding (GTAW) simultaneously, and uses multi‐sensor information fusion technology to fuse the information acquired by the two sensors. The purpose of this paper is to explore the feasibility and effectiveness of multi‐sensor information fusion in pulsed GTAW.

The weld voltage and weld sound information are first acquired by arc sensor and sound sensor, then the features of the two signals are extracted, and the features are fused by weighted mean method to predict the changes of arc length. The weights of each feature are determined by optional distribution method.

The research findings show that multi‐sensor information fusion technology can effectively utilize the information of different sensors and get better result than single sensor.

The arc sensor and sound sensor are first used at the same time to get information about pulsed GTAW and the fusion result shows its advantages over single sensor; this reveals that multi‐sensor fusion technology is a valuable research area in welding process.

Visual simultaneous localization and mapping (SLAM) has limitations such as sensitivity to lighting changes and lower measurement accuracy. The effective fusion of information from multiple modalities to address these limitations has emerged as a key research focus. This study aims to provide a comprehensive review of the development of vision-based SLAM (including visual SLAM) for navigation and pose estimation, with a specific focus on techniques for integrating multiple modalities.

This paper initially introduces the mathematical models and framework development of visual SLAM. Subsequently, this paper presents various methods for improving accuracy in visual SLAM by fusing different spatial and semantic features. This paper also examines the research advancements in vision-based SLAM with respect to multi-sensor fusion in both loosely coupled and tightly coupled approaches. Finally, this paper analyzes the limitations of current vision-based SLAM and provides predictions for future advancements.

The combination of vision-based SLAM and deep learning has significant potential for development. There are advantages and disadvantages to both loosely coupled and tightly coupled approaches in multi-sensor fusion, and the most suitable algorithm should be chosen based on the specific application scenario. In the future, vision-based SLAM is evolving toward better addressing challenges such as resource-limited platforms and long-term mapping.

This review introduces the development of vision-based SLAM and focuses on the advancements in multimodal fusion. It allows readers to quickly understand the progress and current status of research in this field.

In the context of fire incidents within buildings, efficient scene perception by firefighting robots is particularly crucial. Although individual sensors can provide specific types of data, achieving deep data correlation among multiple sensors poses challenges. To address this issue, this study aims to explore a fusion approach integrating thermal imaging cameras and LiDAR sensors to enhance the perception capabilities of firefighting robots in fire environments.

Prior to sensor fusion, accurate calibration of the sensors is essential. This paper proposes an extrinsic calibration method based on rigid body transformation. The collected data is optimized using the Ceres optimization algorithm to obtain precise calibration parameters. Building upon this calibration, a sensor fusion method based on coordinate projection transformation is proposed, enabling real-time mapping between images and point clouds. In addition, the effectiveness of the proposed fusion device data collection is validated in experimental smoke-filled fire environments.

The average reprojection error obtained by the extrinsic calibration method based on rigid body transformation is 1.02 pixels, indicating good accuracy. The fused data combines the advantages of thermal imaging cameras and LiDAR, overcoming the limitations of individual sensors.

This paper introduces an extrinsic calibration method based on rigid body transformation, along with a sensor fusion approach based on coordinate projection transformation. The effectiveness of this fusion strategy is validated in simulated fire environments.

The status of welding process is difficult to monitor because of the intense disturbance during the process. The purpose of this paper is to use multiple sensors to obtain information about the process from different aspects and use multi‐sensor information fusion technology to fuse the information, to obtain more precise information about the process than using a single sensor alone.

Arc sensor, visual sensor, and sound sensor were used simultaneously to obtain weld current, weld voltage, weld pool's image, and weld sound about the pulsed gas tungsten‐arc welding (GTAW) process. Then special algorithms were used to extract the signal features of different information. Fuzzy measure and fuzzy integral method were used to fuse the extracted signal features to predict the penetration status about the welding process.

Experiment results show that fuzzy measure and fuzzy integral method can effectively utilize the information obtained by different sensors and obtain better prediction results than a single sensor.

Arc sensor, visual sensor, and sound sensor are used in pulsed GTAW at the same time to obtain information, and fuzzy measure and fuzzy integral method are used to fuse the different features in welding process for the first time; experiment results show that multi‐sensor information can obtain better results than single sensor, this provides a new method for monitoring welding status and to control the welding process more precisely.

The purpose of this paper is to propose a theoretical framework of applying the Internet of Things (IoT) technologies to the intelligent evacuation protocol in libraries at emergency situations.

The authors conducted field investigations on eight libraries in Wuhan, China, analyzed the characteristics of crowd gathering in libraries and the problems of the libraries’ existing evacuation plans. Therefore, an IoT-based intelligent evacuation protocol in libraries was proposed. Its basic structure consisted of five components: the information base, the protocol base, the IoT sensors, the information fusion system and the intelligent evacuation protocol generation system. In the information fusion system, Dempster–Shafer (D-S) evidence theory was employed as the information fusion algorithm to fuse the multi-sensor information at multiple time points, so as to reduce the uncertainty of disaster prediction. The authors also conducted a case study on the Library L in Wuhan, China. A specific evacuation route was generated for a fire and the crowd evacuation was simulated by the software Patherfind.

The proposed IoT-based evacuation protocol has four distinguishing features: scenario corresponding, precise evacuation, dynamic correction and intelligent decision-making. The case study shows that the proposed protocol is feasible in practice, indicating that the IoT technologies have great potential to be successfully applied to the safety management in libraries.

The software and hardware requirements as well as the Internet network requirements of IoT technologies need to be further discussed.

The proposed IoT-based intelligent evacuation protocol can be widely used in libraries, which is one of the inspirations for the use of IoT technologies in modern constructers.

The application of IoT technologies in libraries is a brand-new topic that has drawn much attention in academia recently. The crowd safety management in libraries is of great significance, and there is little professional literature on it. This paper proposes an IoT-based intelligent evacuation protocol, aiming at improving the safety management in libraries at emergency situations.

Self-localization of an underwater robot using global positioning sensor and other radio positioning systems is not possible, as an alternative onboard sensor-based self-location estimation provides another possible solution. However, the dynamic and unstructured nature of the sea environment and highly noise effected sensory information makes the underwater robot self-localization a challenging research topic. The state-of-art multi-sensor fusion algorithms are deficient in dealing of multi-sensor data, e.g. Kalman filter cannot deal with non-Gaussian noise, while parametric filter such as Monte Carlo localization has high computational cost. An optimal fusion policy with low computational cost is an important research question for underwater robot localization.

In this paper, the authors proposed a novel predictive coding-biased competition/divisive input modulation (PC/BC-DIM) neural network-based multi-sensor fusion approach, which has the capability to fuse and approximate noisy sensory information in an optimal way.

Results of low mean localization error (i.e. 1.2704 m) and computation cost (i.e. 2.2 ms) show that the proposed method performs better than existing previous techniques in such dynamic and unstructured environments.

To the best of the authors’ knowledge, this work provides a novel multisensory fusion approach to overcome the existing problems of non-Gaussian noise removal, higher self-localization estimation accuracy and reduced computational cost.

With the continuous development of online shopping, analyzing the competitiveness of products in the fierce market competition is becoming increasingly crucial to position their own product development. However, the information overload brought by the network development makes it getting difficult to obtain the accurate competitiveness information. Therefore, competitiveness analysis research to combine with the perceived helpfulness study needs urgent solution. Furthermore, deviations exist in the three common methods of perceived helpfulness research. Finally, the traditional information fusion analysis only analyzes the advantages and disadvantages of products in competitiveness analysis without taking account of the competitive environment.

This study puts forward a novel prediction model of perceived helpfulness in conjunction of unsupervised learning and sentiment analysis techniques, to conduct the comparison with pros and cons of congeneric products.

This paper adopts Wilcoxon test to demonstrate the significant rectification of our competitiveness analysis to the traditional methods. It is noted that the positive reviews of the products in this study impact more on product word of mouth and competitiveness than negative ones.

To sum up, the results of this study benefit businesses in locating their dynamic market position with competitors in practice and exploring new method for long-term development strategic planning.