With the increasing demand for product customization, the exponential increase in large-scale and small-batch production orders has yielded new challenges for the capacity of traditional production lines. Due to the information redundancy and complex production process on the production site of the factory, the traditional monitoring system is incapable of meeting the information interaction requirements between the factory management level and the executive level. Meanwhile, the traditional augmented reality (AR) based on image recognition is not suitable in the complex industrial environments. To address the gap, we propose a novel mobile production monitoring system (AIMPMs) with human-in-the-loop control by leveraging the cutting-edge AR and indoor positioning technique. In the proposed system, Ultra-wideband (UWB) and Inertial Measurement Unit (IMU) fusion indoor positioning technology is proposed, which provides accurate indoor positioning information for the production factors in the factory. Subsequently, we build the lightweight indoor map for positioning that can serve as the location reference and path planning, and a nearest-neighbor decision algorithm with double rejection decision (NN-DRD) is proposed to match the positioning features to trigger virtual monitoring information. Finally, the AIMPMs is applied in a hydraulic cylinder factory to verify its enforceability and effectiveness, and the human factors evaluation model and index system are constructed to evaluate two systems, (1) the mobile monitoring system based on positioning information, and (2) the mobile monitoring system based on image recognition, respectively. The experimental results indicate that after the application of AIMPMs, the physiological and mental fatigue of the production personnel is immensely decreased. Therefore, the system realizes a smarter and highly humanized human-machine interaction mode with human-in-the-loop control.

Ultra-wide-band (UWB) positioning is a satisfying indoor positioning technology with high accuracy, low transmission cost, high speed, and strong penetration capacity. However, there remains a lack of systematic study on inevitable and stochastic errors caused by factors originating from the multipath effect (ME), non-line-of-sight interference (NLOSI), and atmospheric interference (AI) in UWB indoor positioning systems. To address this technical issue, this study establishes a dynamic error-propagation model (DEPM) by mainly considering the ME, NLOSI, and AI. First, we analyze the UWB-signal generation principle and spread characteristics used in indoor positioning scenarios. Second, quantization models of the ME, NLOSI, and AI error factors are proposed based on data from related studies. Third, to adapt to various environments, we present a variable-weighted DEPM based on the quantization models above. Finally, to validate the proposed dynamic error-propagation model, UWB-based positioning experiments in an intelligent manufacturing lab were designed and conducted in the form of static and dynamic longitude-tag position measurements. The experimental results showed that the main influencing factors were ME and NLOSI, with a weight coefficient of 0.975, and AI, with a weight coefficient of 0.00025. This study proposes a quantization approach to main error factors to enhance the accuracy and precision of indoor UWB-positioning systems used in intelligent manufacturing areas.

With the increasing applications of UWB indoor positioning technologies in industrial areas, to further enhance the positioning precision, the UWB/IMU combination method (UICM) has been considered as one of the most effective solutions to reduce non-line-of-sight (NLOS) errors. However, most conversional UICMs suffer from a high probability of positioning failure due to uncontrollable and cumulative errors from inertial measuring units (IMU). Hence, to address this issue, we improved the extended Kalman filter (EKF) algorithm of an indoor positioning model based on UWB/IMU tight combination with a double-loop error self-correction. Compared with conventional UICMs, this improved model consists of new modules for fixing time desynchronization, optimizing the threshold setting for UWB ranging, data fusion in NLOS, and double-loop error estimation, sequentially. Further, systematic error controllability analysis proved that the proposed model could satisfy the controllability of UWB indoor positioning systems. To validate this improved UICM, inevitable obstacles and atmospheric interferences were regarded as Gaussian white noises to verify its environmental adaptability. Finally, the experimental results showed that this proposed model outperformed the state-of-the-art UWB-based positioning models with a maximum deviation of 0.232 m (reduced by 83.93% compared to a pure UWB model and 43.14% compared to the conventional UWB/IMU model) and standard deviation of 0.09981 m (reduced by 88.35% compared to a pure UWB model and 22.21% compared to the conventional UWB-IMU model).

Crowd stability analysis is one of research hotspots to alleviate the severe situation of stampede accidents worldwide. Different from the conventional analysis models for crowd stability based on pedestrian density, this study analyses the characteristics of external disturbances and internal obstacle disturbance based on Lyapunov's theory. The critical range of crowd acceleration in crowd evacuation is obtained, a crowd merging acceleration-critical density time delay model is established, and a stability criterion of acceleration vector based on Lyapunov is obtained based on Lyapunov stability analysis. This provides new information for ensuring the stability of crowd movement in public places, assessing the stability of the crowd in the area, and taking reasonable protection and guidance measures prior to instability of a crowd flow.

It is significant to detect abnormal postures of pedestrians in the crowd to crowd stability control. This study locates the joint points of pedestrians based on the pose estimation algorithm OpenPose. After the analysis of 18 nodes and six body parts, the sudden value of node acceleration is obtained, which is compared with the acceleration of the pedestrian’s centre of mass. When there is at least one difference in the direction or acceleration value of the two, it means that the pedestrian has abnormal behaviour. Furthermore, this study analyses the result of comparing the change of z-coordinate value in pedestrian movement with 20% of pedestrian height. These two judgment methods together constitute the dynamic criterion of pedestrian abnormal posture, and judge whether the pedestrian has abnormal behaviour. Compared with the previous dynamic analysis of pedestrian abnormal posture, the accuracy of abnormal posture judgment is improved. This provides a theoretical basis for crowd stability analysis.

Crowd merging is a complex process, and any sudden external or internal disturbance will destroy the stability of the crowd. The occurrence of abnormal behavior will affect the crowd flow process and inevitably affect the stability of the crowd flow system. The position information of the joint points is obtained through the OpenPose algorithm, and the kinematics characteristics of each node are studied. It is judged whether the number of pedestrians in the crowd and the scale of the building scene are greater than the empirical setting value based on engineering statistical data and expert experience. When the number of pedestrians is more than 2,000 and the total area of the passage is more than 2,000 square meters, the appropriate macro-dynamic model is selected. The Aw-Rascle (AR) fluid dynamics model is selected in this study. The joint point information obtained through the OpenPose is combined with the macroscopic fluid dynamics model to construct a macroscopic crowd flow dynamics model based on the pedestrian's abnormal posture.

In public places, it is significant to analyze the stability of the crowd which can support the crowd management and control, and protect the evacuees safely and effectively. The numerical analysis method of system stability based on Lyapunov theory suffers problems that it is difficult to avoid random errors in the initialization of pedestrian density and velocity, as well as cumulative errors due to time increasing, limiting its application. This study adopts a complementary model of theoretical numerical analysis and machine vision with a parallel convolutional neural network (CNN) model. It proposes an approach of stability analysis and closed-loop verification for crowd merging systems. Thereby, this research provides theoretical and methodological support for planning of the functional layout of crowd flow in public crowd-gathering places and the control measures for stable crowd flow.

Over the years, with continuous expansion of the application fields of intelligent video surveillance, technologies related to human gesture recognition have received more attention and become a research hotspot. This paper first shows the architecture of the intelligent surveillance system, and the corresponding computer vision task of human gesture recognition, such as target detection, feature fusion and scene understanding. Then, in order to better understand related technologies, typical methods based on statistics, template and deep learning are summarized, the characteristics and implementation of convolutional neural networks are introduced in detail. Third, some data sets are listed, and we compare existing models in different projects of human body pose estimation from performance indicators. Finally, the flow chart of crowd posture estimation is designed, with the key modules being explained, which helps better understand the mechanism of gesture recognition. Therefore, this paper has certain theoretical value and application significance.

With the improvement and constantly updating of positioning technology, the requirement for positioning accuracy is also getting higher. UWB is a key positioning technology for the complex environment, Firstly, qualitative analysis of all kinds of errors in UWB indoor positioning is enumerated. And based on different fusion algorithms integrated positioning system is introduced. Finally, multi-data fusion positioning error correction method based on UWB and integrated ROS system is put forward. This proposed model has better robust, which can make full advantage of the positioning advantages of UWB, IMU and SLAM sensors, combined with Kalman Filter algorithm to solve the problem of low positioning accuracy of single sensor in complex environment.

In the post-epidemic era, with the resumption of production, orders for large-scale and small-batch production exponentially increase, new challenges to the productivity of the traditional production line have appeared, such as the redundant information, complex production process, which lead the traditional monitoring system cannot satisfy the requirements of rapid information interaction between management and executive layer in the workshop. According to the healthy control during epidemic, physical contact between workers should be reduced. Therefore, Ultra Wide Band(UWB) indoor localization and Augmented Reality(AR) technology are introduced, based on localization information and mobile device camera attitude estimation, an "actively push" mobile monitoring system can be built, which makes the workshop managers timely and completely monitor the manufacturing resources, non-contact production was also ensured. Finally, taking one equipment in a workshop as an example, the virtual visualization models are constructed in Unity software, and the effectiveness of the mobile monitoring system is realized based on UWB indoor localization technology and AR technology.

Pedestrian merging flow in the crowd gathering public places are the common movement nodes of crowd kinematics merging and psychological panic transmission. There are stochastic turbulences, disturbances and density fluctuations in the crowd merging area, with high risk of pedestrian stampede events. Based on the dynamics model of crowd merging, this study considers the psychological characteristics of the escape panic in the normal disaster conditions of the crowd in the cross-passages and the epidemic panic psychological characteristics under the public health events. With the introduction of Shanoon's information entropy theory, panic entropy is applied to measure the degrees of transient panic in the fluid grid area of the crowd, the overall transient disorder of the crowd, and the dynamic relationship with time and space changes. This study comprehensively considers the characteristics of conventional escape panic and epidemic panic, defines protective relaxation factors, forms a dynamic model of escape panic propagation, and provides a scientific theoretical basis for crowd evacuation guidance under COVID-19 epidemic situation.

The centroid is a special point determined by the mass distribution of the object, which is an important intrinsic parameter. The centroid balance is the key technology to ensure the safe operation of the aircraft. However, the limitations of the internal system structure of the aircraft will affect the centroid position, as well as attitude changes and fuel consumption of the aircraft, so it is necessary to develop the corresponding centroid balancing strategy to ensure that the aircraft fully exerts its stability and maneuverability. Firstly, the centroid change curve is drawn based on greedy strategy and mixed integer quadratic programming in this study, then the strategic optimization calculation model that can accurately describe the oil supply problem is established. Finally, using Gurobi to obtain a high-quality feasible solution within the effective time and ensure the centroid balance of the aircraft.

In recent years, digital image processing technology and computer vision technology have been continuously developed and made considerable progress. Based on the image information, through the detection, extraction and recognition operations, the human posture is understood, and abnormal behavior can be recognized in time. In this paper, related work in the abnormal human behavior analysis is first introduced, including many advanced practical projects. Then three methods of human target detection are listed, with the advantages and disadvantages being compared. The detection process is designed according to the inter-frame difference method. Finally, specific realization is proposed from three aspects: area identification, feature extraction, abnormality determination, and the algorithm steps are clearly summarized. Therefore, this study can provide technical support and decision-making guidance for safety management, with a wide range of applications.

As one of the indispensable function layout of pedestrian flow in the crowd gathering public places, cross passages are the common network nodes of crowd kinematics merging and psychology panic propagation. To build the dynamic model of a crowd flow in multi-angle cross, the current micro and macro crowd models are analyzed respectively. The fundamental principles of machine vision are introduced to count pedestrians in a crowd, with a higher precision than the random initialization in conventional numerical crowd models. The typical and often-used crowd merging layouts in multi-angles cross passages are proposed. To increase the adaptability of conventional macro-crowd model, a physical impact matrix is proposed to describe the complex crow merging mechanism in the multi-angle pedestrian passages. Thereby, a novel approach to build the dynamic model of crowd flow in multi-angle cross passages based on machine vision is proposed, which can satisfy the conservation laws of mass, momentum and energy of in crowd merging areas.

Indoor logistics between machining centers is significant to deliver the work-pieces or components just in time a workshop. However, the existing of order disturbance caused by rush order inserting problem, the opaqueness of work in process (WIP) and line-side storage information often increase the difficulty of delivery and even lead to delivery mistakes. To solve the problems above, this paper introduces the augmented reality (AR) technology into a visual management layer. The guidance module diagram of workshop logistics is put forward. Feature-point detection algorithm and description algorithm used for AR is designed. Euclidean distance is defined for two high-dimensional features matching. To validate the approach, an engineering case about the visible delivery of the cylinder shells and piston rod on the CDL pipeline is studied. By using of visualization technology, the production and delivery information of materials can be transmitted to the operators in a comprehensive, real-time and accurate mode, the interaction between operators and equipment can be enhanced, and the entire process from order to production of the workshop products visual management and guidance can be advanced.

A measurement of designing virtual models based on physical entities is helpful for many companies to design their own virtual workshop models for workshop production scheduling and optimization, but a single virtual workshop model cannot reflect the interaction of physical and information objects required for intelligent manufacturing. However, the digital twin workshop is an effective approach to reflect the combination of virtual and physical then to optimize the operation of the workshop production. This study, focusing on the definition and role of digital twin engines in a digital twin workshop, researches on the concept of digital twin workshop, and takes the dual-manipulator cooperation production unit as an example to verify the proposed method for constructing the digital twin workshop.

In the traditional assembly process, it is difficult for operators to quickly understand assembly process information from the instruction manual, which results in inefficient assembly process. Therefore, an assembly guidance system based on augmented reality is about to be proposed in this paper. Also, in view of the problem that geometric models cannot fully express part assembly relationships, this paper proposes to analyze and construct the assembly semantic information model of parts based on the part object model. On the basis of the above semantic information model, the assembly relationship between parts is visually modeled, and these information models are accurately superposed on the real assembly guidance environment by using AR and virtual space modeling technology. Finally, we recursively decompose the parts assembly situation in this paper, and establish the part assembly guidance prototype system to verify the order and accuracy of the part assembly guidance process.