Search results

To realize the smooth interpolation of orientation on robot end-effector, this paper aims to propose a novel algorithm based on the unit quaternion spline curve.

This algorithm combines the spherical linear quaternion interpolation and the cubic B-spline quaternion curve. With this method, a C2-continuous smooth trajectory of multiple teaching orientations is obtained. To achieve the visualization of quaternion curves on a unit sphere, a mapping algorithm between a unit quaternion and a point on the spherical surface is given based on the physical meaning of the unit quaternion.

Finally, the curvature analysis of a practical case shows that the orientation trajectory (OT) constructed by this algorithm satisfied the C2-continuity.

This OT satisfies the requirement of smooth interpolation among multiple orientations on robots in industrial applications.

In recent years, a novel method for computing loadflow orientations and loadpaths, which is based on iterative solutions of non‐linear equations and finite element results, has emerged in the literature. In the present investigation, the prior formulation and approach has been enhanced by deriving explicit expressions for computing loadflow orientations. The new equations produce more accurate loadflow orientations and improve the fidelity of calculated loadpaths. In particular, for a typical loaded plate containing a hole, the density of loadflow lines is also shown to provide accurate values of stress concentration factor. Subsequently, loadflow visualisation for biaxially loaded plates containing non‐optimal and optimal holes is undertaken to identify key features of the stress distributions. It is found that regions of “recirculation” are apparent for non‐optimal hole shapes, whereas no recirculation zones are present for optimal shapes. In general, it is considered that loadflow visualisation is a simple but powerful tool for use by structural designers and analysts.

Robotic ironing needs multidiscipline and requires a quantitative analysis of garment unfolding and ironing motion. This paper investigates the trajectories and orientation of the ironing process where particular geometry is presented in an analytical way. The trajectories produced from this process are analysed and presented with mathematical models to be possibly implemented in robotic automation. This paper further investigates the orientation of iron during the ironing process. It is revealed that the orientation is dependent on the regions of garment and on the closeness to an operator. The orientation is then integrated into the trajectory and presented in a 3D form in which the vertical axis represent the orientation and horizontal axis represent the position. This type of orientation analysis is then used to find similarity in motions to determine the most effective and efficient way of ironing a garment.

Robotic industrial applications are very well established in the manufacturing industry, while they are relatively in their infancy phase in the construction sector. The need for automation in construction is clear especially in repetitive tasks. The excavation process, which is generally critical in most construction projects, is a prime example of such tasks. This paper addresses automation assistance in excavation. The work utilized the robotics approach towards the automation of a typical excavator model, whose structure closely resembled that of an industrial manipulator. A simulation package using Matlab was developed using several embedded design and analysis tools. Emulation was also carried out on the RHINO educational robot to confirm the simulation results. The constructed simulation package offered an integrated environment for trajectory design and analysis for an excavator while addressing the constraints related to the excavator structure, safety and stability, and mode of application.

For efficient trajectory control of industrial robots, a cumbersome computation for inverse kinematics and inverse dynamics is needed, which is usually developed using spatial transformation using Denavit–Hartenberg principle and Lagrangian or Newton–Euler methods, respectively. The model is highly non-linear and needs to deal with uncertainties because of lack of accurate measurement of mechanical parameters, noise and non-inclusion of joint friction, which results in some inaccuracies in predicting accurate torque trajectories. To get a guaranteed closed form solution, the robot designers normally follow Pieper’s recommendation and compromise with the mechanical design. While this may be acceptable for the industrial robots where the aesthetic look is not that important, it is not for humanoid and social robots. To help solve this problem, this study aims to propose an alternative machine learning-based computational approach based on a multi-gated sequence model for finding appropriate mapping between Cartesian space to joint space and motion space to joint torque space.

First, the authors generate sufficient data required for the sequence model, using forward kinematics and forward dynamics by running N number of nested loops, where N is the number of joints of the robot. Subsequently, to develop a learning-based model based on sequence analysis, the authors propose to use long short-term memory (LSTM) and hence, train an LSTM model, the architecture details of which have been discussed in the paper. To make LSTM learning algorithms perform efficiently, the authors need to detect and eliminate redundant features from the data set, which the authors propose to do using an elegant statistical tool called Pearson coefficient.

To validate the proposed model, the authors have performed rigorous experiments using both hardware and simulation robots (Baxter/Anukul robot) available in their laboratory and KUKA simulation robot data set made available from Neural Learning for Robotics Laboratory. Through several characteristic plots, it has been shown that a sequence-based LSTM model of deep learning architecture with non-redundant features could help the robots to learn smooth and accurate trajectories more quickly compared to data sets having redundancy. Such data-driven modeling techniques can change the future course of direction of robotics research for solving the classical problems such as trajectory planning and motion planning for manipulating industrial as well as social humanoid robots.

The present investigation involves development of deep learning-based computation model, statistical analyses to eliminate redundant features, data creation from one hardware robot (Anukul) and one simulation robot model (KUKA), rigorously training and testing separately two computational models (specially configured two LSTM models) – one for learning inverse kinematics and one for learning inverse dynamics problem – and comparison of the inverse dynamics model with the state-of-the-art model. Hence, the authors strongly believe that the present paper is compact and complete to get published in a reputed journal so that dissemination of new ideas can benefit the researchers in the area of robotics.

The purpose of this paper is to conduct a reliable remote manipulation with good contact perception of the remote site. The long-term experience of the authors’ repeatedly confirm that the highest relevance lies in monitoring the wrench acting at a structurally weak point of the work piece rather than monitoring the wrench experienced by the robot end-effector.

The approach followed here is to sense the wrench at the interface of the robot end-effector and the environment. Position and orientation data and environment model are used to arrive at the contact point in real time. The intent of remote contact procedure is understood based on the knowledge of motion trajectory. All the above information is used to develop a wrench transformation to obtain the force diagrams.

The haptic solutions greatly suffer from objectivity, and therefore may result in inconsistency in an operator’s role. Intermediary telepresence through the visual communication of the wrench at the remote site in the form of force diagram provides excellent consistency across the operators and operations. Observing six components of the wrench in separate graphs does not provide on-line error estimate. Force diagrams suggested in the paper are found to be highly effective in perceiving the wrench.

The contact mode operations like assembly, surgery, docking, etc. still suffer due to the lack of easily perceivable wrench visualization. This paper provides solution to such practical issues.

The concept is original, and has evolved steadily over a period of time.

This study aims to provide the theoretical roots, research trajectories and promising research directions of green marketing in marketing and related fields.

This study integrates three bibliometric analyses, i.e. co-citation analysis, historical direct citation analysis and co-occurrence analysis, and a qualitative review to examine the 781 publications that matched the search criteria between January 1991 and December 2021 from the Web of Science (WoS) database.

The research findings show that eleven groups of cited references characterize the theoretical roots of green marketing in marketing and related fields. Besides, the two main research trajectories identified were found to be developed under the impact of prior studies. Moreover, the four research themes concerning this research domain are presented. The results also highlight promising research directions.

Based on the quantitative bibliometric analysis and qualitative literature reviews, this study has provided a comprehensive overview of the current stage of this domain. The study also has underscored an abundance of green marketing literature and revealed the research topics that require further investigations to theoretically and empirically advance the understanding of green marketing.

Some topics about green marketing were recommended for further research. Some practical examples and suggestions are also given in the study.

This study reviewed 781 publications at the intersection of green marketing research domain in marketing and related fields to identify the theoretical roots, research trajectories and research themes and to propose promising research avenues.

Mixed reality is expanding in the industrial market and several companies in various fields are adapting this set of technologies for various purposes, such as optimizing processes, improving the programming tasks and promoting the interactivity of their products with the users, or even improving teaching or training. Robotics is another area that can benefit from these recent technologies. In fact, most of the current and futuristic robotic applications, namely, the areas related to advanced manufacturing tasks (e.g. additive-manufacturing, collaborative robotics, etc.), require new technics to actually perceive the result of several actions, including programming tasks, anticipate trajectories, visualize the motion and related information, interface with programmers and users and several other human–machine interfaces. Consequently, this paper aims to explain a new concept of human–machine interfaces aiming to improve the interaction between advanced users and industrial robotic work cells.

The presented concept uses two different applications (apps) developed to explore the advanced features of the Microsoft HoloLens device. The objectives of the project reported in this paper are to optimize robot paths, just by allowing the advanced user to adjust the selected path through the mixed reality environment, and create new paths, just by allowing the advanced user to insert points in the mixed reality environment, correct them as needed, connect them using a certain type of motion, parametrize them (in terms of velocity, motion precision, etc.) and command them to the robot controller.

The solutions demonstrated in this paper show how mixed reality can be used to allow users, with limited programming experience, to fully use the robotics fields. They also show clearly that the integration of the mixed reality technology in the current robot systems will be a turning point in reducing the complexity for end-users.

There are two challenges in the developed applications. The first relates to the robot tool identification, which is very sensitive to lighting conditions or to very complex robot tools. This can result in positioning errors when the software shows the path in the mixed reality scene. The paper presents solutions to overcome this problem. Another unattended challenge is associated with handling the robot singularities when adjusting or creating new paths. Ongoing work is concentrated in creating mechanisms that prevent the end-user to create paths that contain unreachable points or paths that are not feasible because of bad motion parameters.

This paper demonstrates the utilization of mixed reality device to improve the tasks of programming and commanding manufacturing work cells based on industrial robots [see video in (Pires et al., 2018)]. As the presented devices and robot cells are the basis for Industry 4.0 objectives, this demonstration has a vast field of application in the near future, positively influencing the way complex applications, that require much close cooperation between humans and machines, are thought, planned and built. The paper presents two different applications fully ready to use in industrial environments. These applications are scientific experiments designed to demonstrate the principles and technologies of mixed reality applied to industrial robotics, namely, for improving the programming task.

Although the HoloLens device opens outstanding new areas for robot command and programming, it is still expensive and somehow heavy for everyday use. Consequently, this opens an opportunity window to combine these devices with other mobile devices, such as tablets and phones, building applications that take advantage of their combined features.

The paper presents two different applications fully ready to use in industrial environments. These applications are scientific experiments designed to demonstrate the principles and technologies of mixed reality applied to industrial robotics, namely, for improving the programming task. The first application is about path visualization, i.e. enables the user to visualize, in a mixed reality environment, any path preplanned for the robot cell. With this feature, the advanced user can follow the robot path, identify problems, associate any difficulty in the final product with a particular issue in the robot paths, anticipate execution problems with impact on the final product quality, etc. This is particularly important for not only advanced applications, but also for cases where the robot path results from a CAD package (in an offline fashion). The second application consists of a graphical path manipulation procedure that allows the advanced user to create and optimize a robot path. Just by exploring this feature, the end-user can adjust any path obtained from any programming method, using the mixed reality approach to guide (visually) the path manipulation procedure. It can also create a completely new path using a process of graphical insertion of point positions and paths into the mixed reality scene. The ideas and implementations of the paper are original and there is no other example in the literature applied to industrial robot programming.

The purpose of this paper is to develop a robotic tooth brushing simulator mimicking realistic tooth brushing motions, thereby facilitating greater understanding of the generation of realistic tooth brushing motion for optimal design of toothbrushes.

Tooth brushing motions were measured via a motion capture system. Different motion patterns of brushing were analysed. A series of elliptical motion segments were generated by interpolating ellipse‐like trajectories. Furthermore, a path generation algorithm for brushing simulation was proposed. A path planning system incorporating robot motion control was developed to simulate realistic tooth brushing. The generality and efficiency of the proposed algorithm was demonstrated through simulation and experimental results.

The interpolation of ellipse‐like trajectories can generate elliptical motion segments. Furthermore, realistic tooth brushing can be achieved by integrating the elliptical motion segments into the path generated from the surfaces of teeth. The brushing simulator demonstrated good reproducibility of clinically standardized tooth brushing.

A robotic toothbrush assessment system is a potential application to the robotic tooth brushing simulator by incorporating control of brushing variables, including brushing pressure, speed and temperature.

This study demonstrates the feasibility of using robotic simulation techniques towards improved realistic human tooth brushing motions simulation for optimal design of tooth brushes.

The application of thermal barrier coatings (TBCs) allows aero-engine blades to operate at higher temperatures with higher efficiency. The preparation of the TBCs increases the surface roughness of the blade, which impacts the thermal cycle life and thermal insulation performance of the coating. To reduce the surface roughness of blades, particularly the blades with small size and complex curvature, this paper aims to propose a method for industrial robot polishing trajectory planning based on on-site measuring point cloud.

The authors propose an integrated robotic polishing trajectory planning method using point cloud processing technical. At first, the acquired point cloud is preprocessed, which includes filtering and plane segmentation algorithm, to extract the blade body point cloud. Then, the point cloud slicing algorithm and the intersection method are used to create a preliminary contact point set. Finally, the Douglas–Peucker algorithm and pose frame estimation are applied to extract the tool-tip positions and optimize the tool contact posture, respectively. The resultant trajectory is evaluated by simulation and experiment implementation.

The target points of trajectory are not evenly distributed on the blade surface but rather fluctuate with surface curvature. The simulated linear and orientation speeds of the robot end could be relatively steady over 98% of the total time within 20% reduction of the rest time. After polishing experiments, the coating roughness on the blade surface is reduced dramatically from Ra 7–8 µm to below Ra 1.0 µm. The removal of the TBCs is less than 100 mg, which is significantly less than the weight of the prepared coatings. The blade surface becomes smoothed to a mirror-like state.

The research on robotic polishing of aero-engine turbine blade TBCs is worthwhile. The real-time trajectory planning based on measuring point cloud can address the problem that there is no standard computer-aided drawing model and the geometry and size of the workpiece to be processed differ. The extraction and optimization of tool contact points based on point cloud features can enhance the smoothness of the robot movement, stability of the polishing speed and performance of the blade surface after polishing.