Search

Human-robot collaboration (HRC), as an emerging production mode, has garnered significant attention in the context of the growing smart manufacturing. This study aims to investigate the application of human-robot collaboration in parallel assembly lines and explore its potential for improving productivity, resource utilisation and flexibility. To the best of our knowledge, this is one of the first attempts to consider the combination of parallel assembly line balancing problem (PALBP) and human-robot collaboration. In this study, the parallel assembly line balancing problem with human-robot collaboration (PALBP-HRC) is introduced and characterised. The collaboration mode allows human workers and robots to execute tasks both in parallel and in collaboration. A mixed-integer programming model (MIP) that minimises the cycle time is formulated and a lower bound is proposed. Due to the NP-hardness, we develop an adapted tabu search (TS) algorithm specifically tailored for problem characterisation. Experimental results indicate that the proposed TS algorithm achieves competing performance compared to the MIP and other algorithms, including genetic algorithm (GA) and simulated annealing (SA) algorithm. The comparison between PALBP and PALBP-HRC demonstrates that the integration of human-robot collaboration to parallel assembly line can achieve significant improvements in efficiency and flexibility. [ABSTRACT FROM AUTHOR]

Nowadays industrial robots have been increasingly widely deployed across various sectors and assembly tasks are seen as one of the dominant application fields. The assembly tasks, as the critical process in manufacturing, are still challenging for the robot because of the complex contact state between the robot and the environment (i.e. assembly components). In the assembly task, the robot should switch its controller from a non-contact mode to a contact-rich mode when the contact condition changes, where compliant behaviour is necessary for robustness to uncertain contact and safety of physical interaction. This paper proposes a deep reinforcement learning (DRL) method to achieve such compliance using variable stiffness compliant control. Concretely, a Cartesian compliant controller is built on a virtual dynamics model with a variable non-diagonal stiffness matrix to derive desired motion that reacts with the external force/torque. Upon that, a deep deterministic policy gradient (DDPG)-based agent is deployed to fine-tune such a non-diagonal stiffness matrix. After error-and-trial learning, robots can handle changes in contact state in assembly tasks without any pause and switching its controller mode, thus increasing task efficiency and compliance. The simulation and experiments show that our method allows robots to complete assembly tasks safely and efficiently under noisy observation. [ABSTRACT FROM AUTHOR]

Due to the growing demand for customised products, companies have faced increasing product and process complexity levels. To address this issue, manufacturing processes should become more flexible. One of the most promising technologies to achieve this goal is collaborative robotics (or 'cobots'). In collaborative assembly processes, human and robot combine their skills. However, the co-existence of humans and cobots in the same workspace may influence the operators' perception of assembly complexity. The analysis and control of assembly complexity are crucial to achieving better performances in terms of process quality and operators' well-being. Many qualitative methods have been proposed in the literature to provide a holistic assessment of assembly complexity. This paper proposes a novel method to define a quantitative scale of perceived assembly complexity, based on Thurstone Law of Comparative Judgements. This method was applied to an experimental case-study concerning the assembly of three different products in two modalities (i.e. manual and collaborative). Regression analysis showed that the perceived complexity may be related to the occurrence of process failures and to the perceived workload. The method also proved capable of identifying assembly processes where cobot assistance was helpful, providing process designers with a supporting tool to minimise perceived complexity. [ABSTRACT FROM AUTHOR]

The precise measurement of temperature is crucial in various fields such as biology, medicine, industrial automation, energy management, and daily life applications. While in most scenarios, sensors with a fixed thermal conductivity inevitably mismatch the analogous parameter of the medium being measured, thus causing the distortion and inaccurate detection of original temperature fields. Despite recent efforts on addressing the parameter-mismatch issue, all current solutions are constrained to a fixed working medium, whereas a more universal sensor should function in a variety of scenes. Here, we report a dynamic and reconfigurable thermal sensor capable of highly accurate measurements in diverse working environments. Remarkably, thanks to the highly tunable thermal conductivity of the expanded-plane structure, this sensor works effectively on background mediums with a wide range of conductivity. Such a development greatly enhances the robustness and adaptability of thermal sensors, setting a solid foundation for applications in multi-physical sensing scenarios. [ABSTRACT FROM AUTHOR]

Human–robot collaboration has been incorporated into production and assembly processes to promote system flexibility, changeability and adaptability. However, it poses new challenges to resource allocation and production scheduling due to its intrinsic uncertainty and also the increasing complexity of resources. This paper investigates a stochastic flow shop scheduling problem in the context of human–robot collaboration. The goal is to achieve efficient utilisation of flexible resources including human workers and cobots and take full advantage of human–robot collaboration in production scheduling. A stochastic Cobb–Douglas production function is utilised to evaluate the production efficiency of human–robot collaboration considering instabilities of human performance. A joint chance-constrained programming model is formulated to ensure that the required system performance can be achieved. A CVaR approximation-based approach is proposed to solve the formulated model with mixed-integer variables and a nonconvex constraint. The effectiveness of the formulated model and the efficiency of the proposed solution approach are evaluated via numerical experiments. Computational results show the superiority of our solution approach over three other approaches including Bonferroni approximation, scenario approach and individual chance-constrained programming. [ABSTRACT FROM AUTHOR]

High quality, safe, and reliable batteries are essential for widespread adoption of electric vehicles. Current Li-ion battery pack manufacturing processes rely on manual inspections to ensure electric vehicle battery quality. Such manual quality control is prone to errors, increasing the chances of defective batteries. This is likely to increase safety and reliability concerns in the public imagination, slowing down the adoption of electric vehicles. Furthermore, manual inspection is time-consuming and likely to become a bottleneck in scaling up electric vehicle battery production. A potential solution to address this need for fast and accurate inspection of batteries is the use of machine vision and robotics. In this study, we use digital twin design and simulation to develop a battery module inspection system that uses cobots and machine vision to inspect electric vehicle batteries for defects. Our proposed system can automate visual quality checks that are currently being done by human operators. The proposed cobotic system has been simulated and validated for a variety of battery defects to achieve fast and reliable detection. Since a digital twin of the cobotic inspection workcell has been used, the battery inspection system, as designed and validated, is ready for immediate implementation. [ABSTRACT FROM AUTHOR]

We document significant negative effects of exposure to increased automation at work on household wealth accumulation. Beyond the income and savings channels, we uncover a novel mechanism contributing to the negative wealth effects of automation that arises through the endogenous optimal portfolio decisions of households. We show that households rebalance their financial wealth away from the stock market in response to increased human capital risk induced by pervasive automation, thereby attaining lower wealth levels and relative positions in the wealth distribution. Our evidence suggests that the portfolio channel amplifies the inequality-enhancing effects of increased automation. Authors have furnished an Internet Appendix , which is available on the Oxford University Press Web site next to the link to the final published paper online. [ABSTRACT FROM AUTHOR]

As electric vehicle adoption accelerates and demand increases, the inability to produce batteries in sufficient quantities has emerged as a critical bottleneck in the electric vehicle supply chain. Given the impending climate change crisis, resolving this bottleneck is imperative to accelerate the transition to a zero-emission electric mobility future. One potential solution is the use of robotics for fast and cost-effective assembly of batteries at scale. This study proposes a three-stage digital twin design and analysis method to develop robotic workcells for fast and cost-effective assembly of electric vehicle battery modules. Using digital twin design and simulation, robotic assembly line configurations have been developed for battery module production at different scales. Digital twin analytics was used to evaluate and optimise the proposed robotic battery assembly system for speed and cost. Industrial automation experts were consulted to further improve robotic work cell layouts to minimise investment in robots. Because digital twins of robotic workcells have been used, the configurations of the battery assembly line, as designed and validated, are ready for immediate implementation. For practitioners, this study offers heuristic methods to determine the appropriate assembly line configuration, the required number of robots and humans, for a desired production volume. For researchers, this study outlines promising areas for future investigation. [ABSTRACT FROM AUTHOR]

Collaborative robots (cobots) are a type of Industry 4.0 technology designed to support manufacturing workers and create smart working environments (also called as Industry 5.0). However, little is known about how the use of cobots shapes workers' skills. We analyse this in four types of human-cobot interaction: coexistence, synchronism, cooperation, and collaboration. We examine the implementation of cobots by a leading global provider using a qualitative research based on: (i) analysis of reports regarding the implementation of 200 cobots in 138 companies, (ii) interviews with the team and customers, (iii) six-month follow-up of cobot implementation in a manufacturing plant, and (iv) interviews with two cobot competitors. Our findings demonstrate how each type of human-cobot interaction influences workers' skills in various manufacturing activities. We observe that most companies are in early stages of implementation, focusing on worker substitution. However, we identify a range of effects, including deskilling or reskilling, depending on the type of manufacturing activity analysed. The upskilling effect is particularly evident in the most advanced types of human-cobot interaction, regardless of the company's size. As a main contribution, this paper sheds light on how companies can enhance workers' skills through other levels of interaction between workers and cobots. [ABSTRACT FROM AUTHOR]

With the increased use of collaborative robots, a new production model of the human-robot collaborative hybrid assembly cell (HRCHAC) is becoming a new trend in customised production. Collaborative assembly between workers and robots in assembly cells can significantly increase productivity and improve the well-being of workers once the distribution of tasks and resources is optimised. This paper proposes a new integrated task allocation model to better utilise human-robot collaboration to increase productivity and improve worker well-being. The developed model enables the skills of both workers and robots to be fully utilised while ensuring economic efficiency and the effective protection of workers' physiological and psychological health. First, the product assembly process is decomposed into several assembly tasks, and the characteristics of each task are analysed. Second, a bi-objective mixed-integer planning model is developed with the objectives of minimising unit product assembly time and maximising total task matching. The ergonomics-related objectives are considered in terms of both the physiological and psychological fatigue of the worker, and relevant constraints are established. An improved NSGA-II algorithm is developed to determine the final task allocation scheme. Finally, the proposed method is applied to a real industrial case to verify the effectiveness of the approach. [ABSTRACT FROM AUTHOR]

The U.S. Occupational Safety and Health Administration's (OSHA) most recent estimates show that between 35,000 and 62,000 injuries occur every year due to forklift-related accidents. According to the National Safety Council (NSC) data, approximately 78 fatalities are reported every year. Moreover, manual loading and unloading of heavy items is time-consuming and poses significant risks to workers in small and crowded warehouses. To address these safety and efficiency concerns cost-effectively, an automated robotic forklift prototype was developed. The key features of this industrial robot include full rotational mobility with a zero-degree turning radius, which reduces the time and space required to turn around corners. It can be operated remotely via a mobile phone using Bluetooth or wi-fi. The motion control system, based on the ESP-32 microcontroller, significantly enhances its operational efficiency compared to manual operation. This study evaluates the performance of the robotic forklift prototype, cost-effectiveness, with loading and unloading capabilities as effective solutions to the challenges faced by workers. Additionally, structural analysis using Ansys confirmed that the design can safely withstand forces 60% greater than the intended design load of 50 N. Furthermore, the maximum stress experienced by the fork is 67% below the material yield strength, further demonstrating robustness and reliability. The integration of advanced technology and Eco-friendly design positions this forklift as a viable and sustainable option for improving material handling in various industrial sectors. [ABSTRACT FROM AUTHOR]

This study investigates the potential of deploying a neural network model on an advanced programmable logic controller (PLC), specifically the Finder Opta™, for real-time inference within the predictive maintenance framework. In the context of Industry 4.0, edge computing aims to process data directly on local devices rather than relying on a cloud infrastructure. This approach minimizes latency, enhances data security, and reduces the bandwidth required for data transmission, making it ideal for industrial applications that demand immediate response times. Despite the limited memory and processing power inherent to many edge devices, this proof-of-concept demonstrates the suitability of the Finder Opta™ for such applications. Using acoustic data, a convolutional neural network (CNN) is deployed to infer the rotational speed of a mechanical test bench. The findings underscore the potential of the Finder Opta™ to support scalable and efficient predictive maintenance solutions, laying the groundwork for future research in real-time anomaly detection. By enabling machine learning capabilities on compact, resource-constrained hardware, this approach promises a cost-effective, adaptable solution for diverse industrial environments. [ABSTRACT FROM AUTHOR]

To address several challenges, including low efficiency, significant damage, and high costs, associated with the manual harvesting of Agaricus bisporus, in this study, a machine vision-based intelligent harvesting device was designed according to its agronomic characteristics and morphological features. This device mainly comprised a frame, camera, truss-type robotic arm, flexible manipulator, and control system. The FES-YOLOv5s deep learning target detection model was used to accurately identify and locate Agaricus bisporus. The harvesting control system, using a Jetson Orin Nano as the main controller, adopted an S-curve acceleration and deceleration motor control algorithm. This algorithm controlled the robotic arm and the flexible manipulator to harvest Agaricus bisporus based on the identification and positioning results. To confirm the impact of vibration on the harvesting process, a stepper motor drive test was conducted using both trapezoidal and S-curve acceleration and deceleration motor control algorithms. The test results showed that the S-curve acceleration and deceleration motor control algorithm exhibited excellent performance in vibration reduction and repeat positioning accuracy. The recognition efficiency and harvesting effectiveness of the intelligent harvesting device were tested using recognition accuracy, harvesting success rate, and damage rate as evaluation metrics. The results showed that the Agaricus bisporus recognition algorithm achieved an average recognition accuracy of 96.72%, with an average missed detection rate of 2.13% and a false detection rate of 1.72%. The harvesting success rate of the intelligent harvesting device was 94.95%, with an average damage rate of 2.67% and an average harvesting yield rate of 87.38%. These results meet the requirements for the intelligent harvesting of Agaricus bisporus and provide insight into the development of intelligent harvesting robots in the industrial production of Agaricus bisporus. [ABSTRACT FROM AUTHOR]

Industrial robotic arms are often subject to significant end-effector pose deviations from the target position due to the combined effects of nonlinear deformations such as link flexibility, joint compliance, and end-effector load. To address this issue, a study was conducted on the analysis and compensation of end-position errors in a six-degree-of-freedom robotic arm. The kinematic model of the robotic arm was established using the Denavit–Hartenberg (DH) parameter method, and a rigid–flexible coupled virtual prototype model was developed using ANSYS and ADAMS. Kinematic simulations were performed on the virtual prototype to analyze the variation in end-effector position errors under rigid–flexible coupling conditions. To achieve error compensation, an approach based on an Enhanced Crayfish Optimization Algorithm (ECOA) optimizing a BP neural network was proposed to compensate for position errors. An experimental platform was constructed for error measurement and validation. The experimental results demonstrated that the positioning accuracy after compensation improves by 75.77%, fully validating the effectiveness and reliability of the proposed method for compensating flexible errors. [ABSTRACT FROM AUTHOR]

Field implementations of fully underground sensor networks face many practical challenges that have limited their overall adoption. Power management is a commonly cited issue, as operators are required to either repeatedly excavate batteries for recharging or develop complex underground power infrastructures. Prior works have proposed wireless inductive power transfer (IPT) as a potential solution to these power management issues, but misalignment is a persistent issue in IPT systems, particularly in applications involving moving vehicles or obscured (e.g., underground) coils. This paper presents an automated methodology to sense misalignments and align IPT coils using robotic actuators and sequential Monte Carlo methods. The misalignment of a Class EF inverter-driven IPT system was modeled by tracking changes as its coils move apart laterally and distally. These models were integrated with particle filters to estimate the location of a hidden coil in 3D, given a sequence of sensor measurements. During laboratory tests on a Cartesian robot, these algorithms aligned the IPT system within 1 cm (0.025 coil diameters) of peak lateral alignment. On average, the alignment algorithms required less than four sensor measurements for localization. After laboratory testing, this approach was implemented with an agricultural sensor platform at the Utah Agricultural Experiment Station in Kaysville, Utah. In this implementation, a buried sensor platform was successfully charged using an aboveground, vehicle-mounted transmitter. Overall, this work contributes to the field of underground sensor networks by successfully integrating a self-aligning wireless power delivery system with existing agricultural infrastructure. Furthermore, the alignment strategy presented in this work accomplishes coil misalignment correction without the need for complex sensor or coil architectures. [ABSTRACT FROM AUTHOR]

Automation via robotic systems is becoming widely adopted across many industries, but intelligent autonomy in dynamic environments is challenging to implement due to the difficulty of 3D vision. This paper proposes a novel method that utilizes in-situ 2D image processing to simplify 3D segmentation for robotic workspace detection in industrial applications. Using a TOF sensor mounted on a robotic arm, depth images of the workspace are collected. The algorithm identifies the contour of a table, filters extraneous data points, and converts only relevant data to a 3D pointcloud. This pointcloud is processed to identify the precise location of the workspace with regard to the robot. This method has been shown to be 10% more accurate and over 10,000% faster than a human analyzing the data in a GUI-based software using an octree region-based segmentation algorithm and provides consistent results, only limited by the resolution of the camera itself. [ABSTRACT FROM AUTHOR]

Objectives. Musculoskeletal disorders (MSDs) represent a prevalent global occupational health concern, primarily associated with high biomechanical solicitations, mental workload and work pace. Although cobots have shown promise in reducing risks of MSDs, a question of interest still persists as to how the pace in hybrid human–machine collaboration will affect the operator, in terms of both physical and cognitive health and the production.Methods. This study aimed to analyse the impact of pace on productivity, operators’ posture and mental workload in a human–cobot collaboration. The study, involving 20 participants engaged in a collaborative task with a cobot under three cobot-led paces, assessed productivity rapid upper limb assessment (RULA) scores (posture), dual task performance (cognitive resources) and NASA task load index (NASA-TLX) scores (workload).Results. The findings revealed that an excessively high pace had counterproductive effects, leading to reduced efficiency and increased susceptibility to MSDs, both in terms of physical and mental workloads.Conclusion. In the context of a human–cobot collaboration, it is imperative to adapt the pace to operators’ abilities in order to ensure optimal productivity while preserving their health, emphasizing the need for a balanced approach to pace management in such collaborative work scenarios. [ABSTRACT FROM AUTHOR]

The trajectory planning of manipulators plays a crucial role in industrial applications. This importance is particularly pronounced when manipulators operate in environments filled with obstacles, where devising paths to navigate around obstacles becomes a pressing concern. This study focuses on the environment of frame obstacles in industrial scenes. At present, many obstacle avoidance trajectory planning algorithms struggle to strike a balance among trajectory length, generation time, and algorithm complexity. This study aims to generate path points for manipulators in an environment with obstacles, and the trajectory for these manipulators is planned. The search direction adaptive RRT*Connect (SDA-RRT*Connect) method is proposed to address this problem, which adaptively adjusts the search direction during the search process of RRT*Connect. In addition, we design a path process method to reduce the length of the path and increase its smoothness. As shown in experiments, the proposed method shows improved performances with respect to path length, algorithm complexity, and generation time, compared to traditional path planning methods. On average, the configuration space's path length and the time of generation are reduced by 38.7% and 57.4%, respectively. Furthermore, the polynomial curve trajectory of the manipulator was planned via a PSO algorithm, which optimized the running time of the manipulator. According to the experimental results, the proposed method costs less time during the manipulator's traveling process with respect to other comparative methods. The average reduction in running time is 45.2%. [ABSTRACT FROM AUTHOR]

A pseudo-spectral control algorithm based on adaptive Gauss collocation point reconstruction is proposed to efficiently solve the optimal dynamics control problem of industrial robots. A mathematical model for the kinematic relationship and dynamic optimization control of industrial robots has been established. On the basis of deriving the Legendre–Gauss collocation formula, a two-stage adaptive Gauss collocation strategy for industrial robot dynamics control variables was designed to improve the dynamics optimization control effect of industrial robot by improving the solution efficiency of constrained optimization problems. The results show that compared with the control variable parameterization method and the traditional Gaussian pseudo-spectral method, the proposed dynamic optimal control method based on an adaptive Gaussian point reconstruction algorithm can effectively improve the solving time and efficiency of constrained optimization problems, thereby further enhancing the dynamic optimization control and trajectory tracking effect of industrial robots. [ABSTRACT FROM AUTHOR]

This article presents the fundamental principles of robot accuracy. It characterizes a modular robot, describes the measurement setup, and outlines the methodology for evaluating positioning accuracy across different configurations of the modular robot (four, five, and six modules) under varying loads of 6, 10, and 16 kg. An analysis was conducted on the impact of load changes on four- and five-module configurations, as well as the effect of configuration changes on the robot's performance with 6 and 10 kg loads. The findings indicate that both the number of modules and the load affect positioning accuracy. This article highlights the importance of selecting the optimal configuration based on planned industrial tasks to ensure the highest precision and operational efficiency. [ABSTRACT FROM AUTHOR]

This research work presents an integrated method leveraging Convolutional Neural Networks and Recurrent Neural Networks (CNN-RNN) to enhance the accuracy of predictive maintenance and fault detection in DC motor drives of industrial robots. We propose a new hybrid deep learning framework that combines CNNs with RNNs to improve the accuracy of fault prediction that may occur on a DC motor drive during task processing. The CNN-RNN model determines the optimal maintenance strategy based on data collected from sensors, such as air temperature, process temperature, rotational speed, and so forth. The proposed AI model has the capacity to make highly accurate predictions and detect faults in DC motor drives, thus helping to ensure timely maintenance and reduce operational breakdowns. As a result, comparative analysis reveals that the proposed framework can achieve higher accuracy than the current existing method of combining CNN with Long Short-Term Memory networks (CNN-LSTM) as well as other CNNs, LSTMs, and traditional methods. The proposed CNN-RNN model can provide early fault detection for motor drives of industrial robots with a simpler architecture and lower complexity of the model compared to CNN-LSTM methods, which can enable the model to process faster than CNN-LSTM. It effectively extracts dynamic features and processes sequential data, achieving superior accuracy and precision in fault diagnosis, which can make it a practical and efficient solution for real-time fault detection in motor drive control systems of industrial robots. [ABSTRACT FROM AUTHOR]

Human hands have over 20 degrees of freedom, enabled by a complex system of bones, muscles, and joints. Hand differences can significantly impair dexterity and independence in daily activities. Accurate assessment of hand function, particularly digit movement, is vital for effective intervention and rehabilitation. However, current clinical methods rely on subjective observations and limited tests. Smart gloves with inertial measurement unit (IMU) sensors have emerged as tools for capturing digit movements, yet their sensor accuracy remains underexplored. This study developed and validated an IMU-based smart glove system for measuring finger joint movements in individuals with hand differences. The glove measured 3D digit rotations and was evaluated against an industrial robotic arm. Tests included rotations around three axes at 1°, 10°, and 90°, simulating extension/flexion, supination/pronation, and abduction/adduction. The IMU sensors demonstrated high accuracy and reliability, with minimal systematic bias and strong positive correlations (p > 0.95 across all tests). Agreement matrices revealed high agreement (<1°) in 24 trials, moderate (1–10°) in 12 trials, and low (>10°) in only 4 trials. The Root Mean Square Error (RMSE) ranged from 1.357 to 5.262 for the 90° tests, 0.094 to 0.538 for the 10° tests, and 0.129 to 0.36 for the 1° tests. Likewise, mean absolute error (MAE) ranged from 0.967 to 4.679 for the 90° tests, 0.073 to 0.386 for the 10° tests, and 0.102 to 0.309 for the 1° tests. The sensor provided precise measurements of digit angles across 0–90° in multiple directions, enabling reliable clinical assessment, remote monitoring, and improved diagnosis, treatment, and rehabilitation for individuals with hand differences. [ABSTRACT FROM AUTHOR]

Facing the system uncertainties caused by unmodeled dynamics and unpredictable external disturbances, the robot position control for meeting the high-performance control requirements on higher accuracy and faster beat is vital for many industrial applications, such as welding and laser cutting tasks. This work aims to cope with the problem of precise and fast position tracking for robot manipulators with an effective and safe control scheme. Specifically, a discrete-time super-twisting observer (STO) is integrated into the scheme to estimate the uncertain dynamics (e.g. unmodeled dynamics and external disturbances) in the feedforward compensation part of the dynamics. Subsequently, a discrete-time fast terminal sliding mode controller (FTSMC) dominates the robot control to guarantee fast convergence of the position tracking error. The significant improvement of the proposed method with respect to other discrete-time sliding mode control approaches lies in that it is capable of alleviating the chattering-like problem, achieving a fast convergence and improving the robustness of sliding mode control against uncertain dynamics. To illustrate the effectiveness of the presented control scheme, several experiments on a six degrees of freedom (6DoF) robot manipulator are provided. [ABSTRACT FROM AUTHOR]

ABSTRACT A robust approximate constraint‐following control (RACC) approach is proposed in this article for collaborative robots with inequality constraints. The trajectory‐following control and boundary control of the robot are investigated. First, an explicit constraint equation for the collaborative robot system is established based on the Udwadia–Kalaba (U‐K) theory. Second, due to the monotone unbounded property of the tangent function, a special function is constructed to transform the joint output angles of the constrained robot into unconstrained state variables, and a new form of the robot constraint equation is obtained. Through this transformation, the joint motion of the robot will always be confined to specified angles and follow the desired trajectory. The constraint equation ensures the safety of the robot at the algorithmic level and innovatively solves the control problem of the equality and inequality of the robot's motion. According to theoretical analysis, the control approach can deal with uncertainty and satisfy both uniform boundedness (UB) and uniform ultimate boundedness (UUB) requirements. Finally, based on the rapid controller prototype CSPACE and a two‐degree‐of‐freedom collaborative robot platform, experimental verification is carried out. Numerical simulation and experimental results demonstrate that the proposed RACC approach with state transformation exhibits significant advantages in trajectory tracking performance and safety for collaborative robots. [ABSTRACT FROM AUTHOR]

In the construction industry, where work environments are complex, unstructured and often dangerous, the implementation of human–robot collaboration (HRC) is emerging as a promising advancement. This underlines the critical need for intuitive communication interfaces that enable construction workers to collaborate seamlessly with robotic assistants. This study introduces a conversational virtual reality (VR) interface integrating multimodal interaction to enhance intuitive communication between construction workers and robots. By integrating voice and controller inputs with the robot operating system (ROS), building information modeling (BIM), and a game engine featuring a chat interface powered by a large language model (LLM), the proposed system enables intuitive and precise interaction within a VR setting. Evaluated by 12 construction workers through a drywall installation case study, the proposed system demonstrated its low workload and high intuitiveness and ease of use with succinct command inputs. The proposed multimodal interaction system suggests that such technological integration can substantially advance the integration of robotic assistants in the construction industry. [ABSTRACT FROM AUTHOR]

Critical global challenges, such as climate change and the insufficient availability of resources, mean that it is a pivotal time to make cities more intelligent, efficient, and sustainable in a drive towards a net-zero carbon future. This requires intelligent, interactive, and responsive structural health monitoring (SHM) to assure the longevity and safety of ageing infrastructure. Drones have the potential to revolutionise SHM. Drone-based SHM (as a potential fly-by technique) involves equipping drones with various sensors, or using inbuilt sensors, to capture data and images of structures from different angles and perspectives. The data is then processed and analysed to facilitate accurate assessment of the structure's health and early diagnosis of damage. Although the use of fly-by is relatively new, the speedy advances in various technologies that could be integrated with it, such as computer vision with artificial intelligence, deep learning, and links to digital twins, put these systems on the verge of a potential breakthrough. This paper provides an overview of fly-by SHM technique using both scientometric and qualitative systematic literature review processes, in order to provide a distinct understanding of the state of the art of research. As an original contribution, our research identified four main clusters of research within the field of fly-by SHM: (1) the application of UAV-enabled vision-based monitoring; (2) the integration of drones, advanced sensor technologies, and artificial intelligence; (3) drone-based SHM integrating modal analysis, energy harvesting, and deep learning; and (4) automation and robotics in drone-based SHM. The paper highlights the integration of new technologies such as artificial intelligence, machine learning, and sensors with the fly-by technique for SHM, identifies the gaps in current fly-by SHM research, and suggests new directions for research. [ABSTRACT FROM AUTHOR]

This paper presents a map-free navigation approach for industrial automatic mobile robots (AMRs), designed to ensure computational efficiency, cost-effectiveness, and adaptability. Utilizing deep reinforcement learning (DRL), the system enables real-time decision-making without fixed markers or frequent map updates. The central contribution is the Heuristic Dense Reward Shaping (HDRS), inspired by potential field methods, which integrates domain knowledge to improve learning efficiency and minimize suboptimal actions. To address the simulation-to-reality gap, data augmentation with controlled sensor noise is applied during training, ensuring robustness and generalization for real-world deployment without fine-tuning. Training results underscore HDRS's superior convergence speed, training stability, and policy learning efficiency compared to baselines. Simulation and real-world evaluations establish HDRS-DRL as a competitive alternative, outperforming traditional approaches, and offering practical applicability in industrial settings. • Map-free, DRL-based navigation architecture ensures real-time operational efficiency. • Heuristic dense reward shaping improves learning and navigation performance. • Data augmentation with noise strengthens model robustness for direct AMR deployment. [ABSTRACT FROM AUTHOR]

Due to their precision, compact size, and high torque transfer, Rotate vector (RV) reducers are becoming more popular in industrial robots. However, repetitive operations and varying speed conditions mean that these components are prone to mechanical failure. Therefore, it is important to develop effective health monitoring (HM) strategies. Traditional approaches for HM, including those using vibration and acoustic emission sensors, encounter such challenges as noise interference, data inconsistency, and high computational costs. Deep learning-based techniques, which use current electrical data embedded within industrial robots, address these issues, offering a more efficient solution. This research provides transfer learning (TL) models for the HM of RV reducers, which eliminate the need to train models from scratch. Fine-tuning pre-trained architectures on operational data for the three different reducers of health conditions, which are healthy, faulty, and faulty aged, improves fault classification across different motion profiles and variable speed conditions. Four TL models, EfficientNet, MobileNet, GoogleNet, and ResNET50v2, are considered. The classification accuracy and generalization capabilities of the suggested models were assessed across diverse circumstances, including low speed, high speed, and speed fluctuations. Compared to the other models, the proposed EfficientNet model showed the most promising results, achieving a testing accuracy and an F1-score of 98.33% each, which makes it best suited for the HM of robotic reducers. [ABSTRACT FROM AUTHOR]

Copyright of Bearing is the property of Bearing Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

There is a growing interest in whether social robots, which are embodied and exhibit human-like behaviour, can be used for teaching and learning. Still, very few studies focus on the teacher's role. This study focuses on how a teacher acted in a learning-by-teaching activity with 20 children. In this small-scale field experiment, the teacher's interactions and teaching actions were observed when the teacher scaffolded a learning activity where children played a collaborative digital mathematics game to strengthen their mathematical reasoning and conceptual understanding of arithmetic. When playing, the children were acting as tutors for a tutee, according to the learning-by-teaching principle. In one scenario, the tutee was a younger child; in the other, the tutee was a social robot. Twenty 30-minute game-playing sessions are observed, video-recorded, and transcribed. The study explores the teacher's interactions and teaching actions in the two scenarios and discusses the results from the perspective of the teacher's role, social norms, and teacher digital competence. The interaction and thematic analyses show similarities and characteristic differences in the teacher's interaction patterns in the two scenarios. The teaching actions are similar on a structural level and differ regarding the types and distribution of teaching actions. In the child-child scenario, the teacher directs most teaching actions to both players, and the actions are didactic (mathematical) scaffolding. In contrast, in the child-robot scenario, the teacher only addresses the tutor, and the scaffolding is socially oriented. Implications for a teaching practice involving social robots as learning companions are discussed regarding teachers' presence and participation, types of social robot knowledge that go beyond digital competence, and new challenges introduced by using social robots as learning companions in the classroom. The study contributes new insights into the teacher's role and actions when teaching with a social robot in a collaborative learning situation, which is relevant for educational research and teaching practice. [ABSTRACT FROM AUTHOR]