Search results

Automatic robots can improve the efficiency of liquefied petroleum gas (LPG) tank inspection and maintenance, but it is difficult to achieve high-precision spatial positioning and navigation on tank surfaces. The purpose of this paper is to develop a spatial positioning robotic system for tank inspection. The robot can accurately identify and track weld paths. The positioning system can complete robot’s spatial positioning on tank surfaces.

A tank inspection robot with curvature-adaptive transmission mechanisms is designed in this study. A weld path recognition method based on deep learning is proposed to accurately identify and extract weld paths. Integrated multiple sensors, the positioning system is developed to improve the robot’s spatial positioning accuracy. Experiments are conducted on a cylindrical tank to test weld seam tracking accuracy and spatial positioning performance of the robotic system. The practicality of the robotic system is then verified in field tests.

The robot can accurately identify and track weld seams with a maximum drift angle of 4° and a maximum offset distance of ±30 mm. The positioning system has excellent positioning accuracy and stability. The maximum angle and height errors are 3° and 0.08 m, respectively.

The positioning system can improve the autonomous performance of inspection robots and solve the problems of weld path recognition and spatial positioning. Application of the robotic system can promote the automatic inspection and maintenance of LPG tanks.

Stereo vision technique simulates the function of the human eyes to observe the world, which can be used to compute the spatial information of weld seam in the robot welding field. It is a typical kind of application to fix two cameras on the end effector of robot when stereo vision is used in intelligent robot welding. In order to analyse the effect of vision system configuration on vision computing, an accuracy analysis model of vision computing is constructed, which is a good guide for the construction and application of stereo vision system in welding robot field.

A typical stereo vision system fixed on welding robot is designed and constructed to compute the position information of spatial seam. A simplified error analysis model of the two arbitrary putting cameras is built to analyze the effect of sensors' structural parameter on vision computing accuracy. The methodology of model analysis and experimental verification are used in the research. And experiments related with image extraction, robot movement accuracy is also designed to analyze the effect of equipment accuracy and related processed procedure in vision technology.

Effect of repeatability positioning accuracy and TCP calibration error of welding robot for visual computing are also analyzed and tested. The results show that effect of the repeatability on computing accuracy is not bigger than 0.3 mm. However, TCP affected the computing accuracy greatly, when the calibrated error of TCP is bigger than 0.5, the re‐calibration is very necessary. The accuracy analysis and experimental technique in this paper can guide the research of three‐dimensional information computing by stereo vision and improve the computed accuracy.

The accuracy of seam position information is affected by many interactional factors, the systematic experiments and a simplified error analysis model are designed and established, the main factors such as the sensor's configurable parameters, the accuracy of arc welding robot and the accuracy of image recognition, are included in the model and experiments. The model and experimental method are significant for design of visual sensor and improvement of computing accuracy.

The purpose of this paper is to describe how to measure workplace configuration, show its application in, and the results of, a field experiment aimed at improving collaborative knowledge work and identify and discuss larger problems involved with research on workplace configuration.

The paper reviews recent thought on density, proximity and problems with evaluating configuration. It then describes a method of spatial network analysis used in a field experiment involving the reconfiguration of a workplace. This is followed with a discussion of recent research on knowledge work from economic geography.

It was found that, instead of increasing it, the reconfigured workplace decreased collaborative activity. The spatial network analysis shows how this occurred.

This method of spatial network analysis, when used carefully, is a robust technique for analyzing and comparing spatial configuration. Further research needs to address the links between spatial proximity and information and communications technologies as well as the relation of types of knowledge bases and associated forms of proximity that can stimulate collaboration.

While spatial network analysis methods are not do‐it‐yourself tools, corporate real estate managers should employ them, especially in larger‐scale projects, before committing to final workplace designs. They can also use them to identify and map best spatial patterns (like best practices) to identify strategic spatial patterns.

This appears to be the first rigorous application of spatial network methods in a field experiment involving a real workplace. The paper shows the method can clearly extract and discriminate spatial network patterns underlying configurations and relate them both quantitatively and graphically to employee evaluations of collaborative performance. It introduces concepts of comparative knowledge bases that need to be understood in determining the types of collaboration needed in a workplace.

The possibility of integrating building information in an augmented reality (AR) environment provides an effective solution to all phases of a building's lifecycle. This paper explores the integration of building information modelling (BIM) and AR to effectively visualise building information models in an AR environment and evaluates the currently available AR tools.

A BIM model of a selected office room was created and superimposed to the actual physical space using two different AR devices and four different AR applications. The superimposing techniques, accuracy and the level of information that can be visualised were then investigated by performing a walk-through analysis.

From the investigation, it can be concluded that model positioning can be inaccurate depending on the superimposing method used and the AR device. Moreover, using the currently available techniques, only static building information can be superimposed and visualised in AR, showing a need to integrate data from Internet of Things (IoT) sensors into the current BIM-AR processes to allow visualisation of accurate and high-quality operational building information.

A practical process and method for visualising and superimposing BIM models in an AR environment have been described. Recommendations to improve superimposing accuracy are provided. The assessment of type, quality and level of detail that can be visualised indicates the areas that need improvement to increase the effectiveness of building information's visualisation in AR.

The paper aims to introduce an intelligent recognition system for viewpoint variations of gait and speech. It proposes a convolutional neural network-based capsule network (CNN-CapsNet) model and outlining the performance of the system in recognition of gait and speech variations. The proposed intelligent system mainly focuses on relative spatial hierarchies between gait features in the entities of the image due to translational invariances in sub-sampling and speech variations.

This proposed work CNN-CapsNet is mainly used for automatic learning of feature representations based on CNN and used capsule vectors as neurons to encode all the spatial information of an image by adapting equal variances to change in viewpoint. The proposed study will resolve the discrepancies caused by cofactors and gait recognition between opinions based on a model of CNN-CapsNet.

This research work provides recognition of signal, biometric-based gait recognition and sound/speech analysis. Empirical evaluations are conducted on three aspects of scenarios, namely fixed-view, cross-view and multi-view conditions. The main parameters for recognition of gait are speed, change in clothes, subjects walking with carrying object and intensity of light.

The proposed CNN-CapsNet has some limitations when considering for detecting the walking targets from surveillance videos considering multimodal fusion approaches using hardware sensor devices. It can also act as a pre-requisite tool to analyze, identify, detect and verify the malware practices.

This research work includes for detecting the walking targets from surveillance videos considering multimodal fusion approaches using hardware sensor devices. It can also act as a pre-requisite tool to analyze, identify, detect and verify the malware practices.

This proposed research work proves to be performing better for the recognition of gait and speech when compared with other techniques.

The spatial conflicts and congestion of construction resources are challenges that lead to the reduction in efficiency. The purpose of this paper is to enable users to detect and resolve workspace conflicts by implementing four resolution strategies in a five-dimensional (5D) CAD model. In addition to resolving conflicts, the model should be able to optimize time and cost of the projects. In other words, three variables of spatial conflicts, time and cost of project are considered simultaneously in the proposed model to find the optimum solution.

In the first step, a 5D simulation model is developed that includes time, cost and geometrical information of a project. Then, time-cost trade-off analysis was carried out to distinguish optimum schedule. The schedule was imported to the 5D CAD model to detect spatial conflicts. Finally, a novel algorithm was implemented to solve identified conflicts while imposing minimum project’s time and cost. Several iterations are performed to resolve all clashes using conflict resolution algorithm and visual simulation model.

The proposed methodology in this research was applied to a real case. Results showed that in comparison to the normal and initial schedule with 19 conflicts, the finalized schedule has no conflict, while time and cost of the project are both reduced.

Implementing the proposed methodology in construction projects requires proper technical basis in this field. In this regard, the executive user should have a proper understanding of the principles, concepts and tools of building information modeling and have project management knowledge. Also, the implementation conditions of the basic model requires the determination of the construction methods, estimated volumes of working items, scheduling and technical specification. The designed methodology also has two limitations regarding to its implementation. The first is the fact that strategies should be applied manually to the schedule. The other one pertains to the number of strategies used in the research. Four strategies have been used in the conflict resolution algorithm directly and the two others (spatial divisibility and activities breakdown strategies) have been used as default strategies in the visual simulation model. Since the unused strategies including the changing of construction method and the activity resources are subjective and depend upon the planner and project manager’s personal opinion, the authors have avoided using them in this research.

The method proposed in this research contributes the coordination of the working teams at the planning and execution phases of the project. In fact, the best location and work direction for each working team is presented as a schedule, so that the space conflict may not come about and the cost can be minimized. This visual simulation not only deepens the planners’ views about the executive barriers and the spatial conditions of the worksite, it also makes the construction engineers familiar on a daily basis with their executive scope. Therefore, it considerably improves the interactions and communication of the planning and construction teams. Another advantage and application of this methodology is the use of initial and available projects’ documents including the schedule and two-dimensional drawings. The integration of these basic documents in this methodology helps identify the spatial conflicts efficiently. To achieve this, the use of the existing and widely-used construction tools has facilitated the implementation of the methodology. Using this system, planners have applied the strategies in an order of priority and can observe the results of each strategy visually and numerically in terms of time, cost and conflicts. This methodology by providing the effective resolution strategies guides the practitioner to remove conflicts while optimum time and cost are imposed to project.

Contrary to the previous models that ignore cost, the proposed model is a 5D visual simulation model, which considers the variable of cost as a main factor for conflict identification and resolution. Moreover, a forward-pass approach is introduced to implement resolution strategies that are novel compared to other investigations.

Intelligent and connected vehicle technology is in the ascendant. High-level autonomous driving places more stringent requirements on the accuracy and reliability of environmental perception. Existing research works on multitarget tracking based on multisensor fusion mostly focuses on the vehicle perspective, but limited by the principal defects of the vehicle sensor platform, it is difficult to comprehensively and accurately describe the surrounding environment information.

In this paper, a multitarget tracking method based on roadside multisensor fusion is proposed, including a multisensor fusion method based on measurement noise adaptive Kalman filtering, a global nearest neighbor data association method based on adaptive tracking gate, and a Track life cycle management method based on M/N logic rules.

Compared with fixed-size tracking gates, the adaptive tracking gates proposed in this paper can comprehensively improve the data association performance in the multitarget tracking process. Compared with single sensor measurement, the proposed method improves the position estimation accuracy by 13.5% and the velocity estimation accuracy by 22.2%. Compared with the control method, the proposed method improves the position estimation accuracy by 23.8% and the velocity estimation accuracy by 8.9%.

A multisensor fusion method with adaptive Kalman filtering of measurement noise is proposed to realize the adaptive adjustment of measurement noise. A global nearest neighbor data association method based on adaptive tracking gate is proposed to realize the adaptive adjustment of the tracking gate.

The purpose of this paper is to present a follow-the-leader motion strategy for multi-section continuum robots, which aims to make the robot have the motion ability in a confined environment and avoid a collision.

First, the mechanical design of a multi-section continuum robot is introduced and the forward kinematic model is built. After that, the follow-the-leader motion strategy is proposed and the differential evolution (DE) algorithm for calculating optimal posture parameters is presented. Then simulations and experiments are carried out on a series of predefined paths to analyze the performance of the follow-the-leader motion.

The follow-the-leader motion can be well performed on the continuum robots this study proposes in this research. The experimental results show that the deviation from the path is less than 9.7% and the tip error is no more than 15.6%.

Currently, the follow-the-leader motion is affected by the following factors such as gravity and continuum robot design. Furthermore, the position error is not compensated under open-loop control. In future work, this paper will improve the accuracy of the robot and introduce a closed-loop control strategy to improve the motion accuracy.

The main contribution of this paper is to present an algorithm to generate follow-the-leader motion of the continuum robot based on DE. This method is suitable for solving new arrangements in the process of following a nonlinear path. Then, it is expected to promote the engineering application of the continuum robot.

This work analyzes a pharmaceutical supply chain (PSC) in terms of supply chain visibility (SCV). The current good distribution practice (GDP) guideline demands increased visibility from firms. The purpose of this paper is to propose a solution for SCV enhancements based on automatic identification (Auto-ID) technologies.

The authors qualitatively analyze data from ten case studies of actors in a PSC. A review of Auto-ID technologies supports the derivation of solutions to enhance SCV.

This work shows that the functionalities of Auto-ID technologies offered by current practical monitoring solutions and challenges created by the GDP guideline necessitate further SCV enhancements. To enhance SCV, the authors propose three solutions: securPharm with passive radio frequency identification tags, transport containers with sensor nodes, and an SCV dashboard.

This study is limited to a PSC in Germany and is therefore not intended to be exhaustive. Thus, the results serve as a foundation for further analyses.

This study provides an overview of the functionality of Auto-ID technologies. In juxtaposition with the influence of the GDP guideline, the use of our Auto-ID-based solutions can help to enhance SCV.

This work analyzes a PSC in Germany, with consideration given to the influence of current legislation. Based on a multiple-case-study design, the authors derive three Auto-ID-based solutions for enhancing SCV.

Obstacle and wind field are common environmental factors for mini unmanned helicopter (MUH) flight. This paper aims to develop a trajectory planning approach guiding MUH to avoid static and dynamic obstacles and to fly in steady uniform or boundary-layer wind field.

An optimal control model including a nonlinear flight dynamics model and a cubic obstacle model is established for MUH trajectory planning. Radau pseudospectral method is used to generate the optimal trajectory.

The approach can plan reasonable obstacle-avoiding trajectories in obstacle and windy environments. The simulation results show that high-speed wind fields increase the flight time and fluctuation of control inputs. If boundary-layer wind field exists, the trajectory deforms significantly and gets closer to the ground to escape from the strong wind.

The key innovations in this paper include a cubic obstacle model which is straightforward and practical for trajectory planning and MUH trajectory planning in steady uniform wind field and boundary-layer wind field. This study provides an efficient solution to the trajectory planning for MUH in obstacle and windy environments.