Search results

The purpose of this research is to achieve multi-task autonomous driving by adjusting the network architecture of the model. Meanwhile, after achieving multi-task autonomous driving, the authors found that the trained neural network model performs poorly in untrained scenarios. Therefore, the authors proposed to improve the transfer efficiency of the model for new scenarios through transfer learning.

First, the authors achieved multi-task autonomous driving by training a model combining convolutional neural network and different structured long short-term memory (LSTM) layers. Second, the authors achieved fast transfer of neural network models in new scenarios by cross-model transfer learning. Finally, the authors combined data collection and data labeling to improve the efficiency of deep learning. Furthermore, the authors verified that the model has good robustness through light and shadow test.

This research achieved road tracking, real-time acceleration–deceleration, obstacle avoidance and left/right sign recognition. The model proposed by the authors (UniBiCLSTM) outperforms the existing models tested with model cars in terms of autonomous driving performance. Furthermore, the CMTL-UniBiCL-RL model trained by the authors through cross-model transfer learning improves the efficiency of model adaptation to new scenarios. Meanwhile, this research proposed an automatic data annotation method, which can save 1/4 of the time for deep learning.

This research provided novel solutions in the achievement of multi-task autonomous driving and neural network model scenario for transfer learning. The experiment was achieved on a single camera with an embedded chip and a scale model car, which is expected to simplify the hardware for autonomous driving.

This paper aim to solve the problem of low assembly success rate for 3c assembly lines designed based on classical control algorithms due to inevitable random disturbances and other factors,by incorporating intelligent algorithms into the assembly line, the assembly process can be extended to uncertain assembly scenarios.

This work proposes a reinforcement learning framework based on digital twins. First, the authors used Unity3D to build a simulation environment that matches the real scene and achieved data synchronization between the real environment and the simulation environment through the robot operating system. Then, the authors trained the reinforcement learning model in the simulation environment. Finally, by creating a digital twin environment, the authors transferred the skill learned from the simulation to the real environment and achieved stable algorithm deployment in real-world scenarios.

In this work, the authors have completed the transfer of skill-learning algorithms from virtual to real environments by establishing a digital twin environment. On the one hand, the experiment proves the progressiveness of the algorithm and the feasibility of the application of digital twins in reinforcement learning transfer. On the other hand, the experimental results also provide reference for the application of digital twins in 3C assembly scenarios.

In this work, the authors designed a new encoder structure in the simulation environment to encode image information, which improved the model’s perception of the environment. At the same time, the authors used the fixed strategy combined with the reinforcement learning strategy to learn skills, which improved the rate of convergence and stability of skills learning. Finally, the authors transferred the learned skills to the physical platform through digital twin technology and realized the safe operation of the flexible printed circuit assembly task.

Most manufacturing plants choose the easy way of completely separating human operators from robots to prevent accidents, but as a result, it dramatically affects the overall quality and speed that is expected from human–robot collaboration. It is not an easy task to ensure human safety when he/she has entered a robot’s workspace, and the unstructured nature of those working environments makes it even harder. The purpose of this paper is to propose a real-time robot collision avoidance method to alleviate this problem.

In this paper, a model is trained to learn the direct control commands from the raw depth images through self-supervised reinforcement learning algorithm. To reduce the effect of sample inefficiency and safety during initial training, a virtual reality platform is used to simulate a natural working environment and generate obstacle avoidance data for training. To ensure a smooth transfer to a real robot, the automatic domain randomization technique is used to generate randomly distributed environmental parameters through the obstacle avoidance simulation of virtual robots in the virtual environment, contributing to better performance in the natural environment.

The method has been tested in both simulations with a real UR3 robot for several practical applications. The results of this paper indicate that the proposed approach can effectively make the robot safety-aware and learn how to divert its trajectory to avoid accidents with humans within the workspace.

The method has been tested in both simulations with a real UR3 robot in several practical applications. The results indicate that the proposed approach can effectively make the robot be aware of safety and learn how to change its trajectory to avoid accidents with persons within the workspace.

This paper provides a novel collision avoidance framework that allows robots to work alongside human operators in unstructured and complex environments. The method uses end-to-end policy training to directly extract the optimal path from the visual inputs for the scene.

In this research, the authors demonstrate the advantage of reinforcement learning (RL) based intrusion detection systems (IDS) to solve very complex problems (e.g. selecting input features, considering scarce resources and constrains) that cannot be solved by classical machine learning. The authors include a comparative study to build intrusion detection based on statistical machine learning and representational learning, using knowledge discovery in databases (KDD) Cup99 and Installation Support Center of Expertise (ISCX) 2012.

The methodology applies a data analytics approach, consisting of data exploration and machine learning model training and evaluation. To build a network-based intrusion detection system, the authors apply dueling double deep Q-networks architecture enabled with costly features, k-nearest neighbors (K-NN), support-vector machines (SVM) and convolution neural networks (CNN).

Machine learning-based intrusion detection are trained on historical datasets which lead to model drift and lack of generalization whereas RL is trained with data collected through interactions. RL is bound to learn from its interactions with a stochastic environment in the absence of a training dataset whereas supervised learning simply learns from collected data and require less computational resources.

All machine learning models have achieved high accuracy values and performance. One potential reason is that both datasets are simulated, and not realistic. It was not clear whether a validation was ever performed to show that data were collected from real network traffics.

The study provides guidelines to implement IDS with classical supervised learning, deep learning and RL.

The research applied the dueling double deep Q-networks architecture enabled with costly features to build network-based intrusion detection from network traffics. This research presents a comparative study of reinforcement-based instruction detection with counterparts built with statistical and representational machine learning.

Although end‐user computing (EUC) training has received significant attention among academics and practitioners, the effective transfer of trained EUC skills is a relatively neglected issue. Analysis of factors affecting the EUC transfer process will aid in understanding and improving training transfer. Hence, the purpose of this paper is to underscore key trainee characteristics and facets of the work environment that influence EUC training transfer.

The theoretical framework includes prior computer experience, computer anxiety, computer self‐efficacy, pre‐training motivation and perceived job utility as significant trainee factors influencing the EUC transfer process. In addition, the model includes supervisory support as an important constituent of the EUC transfer process.

The model highlights the mediating roles of computer self‐efficacy and pre‐training motivation in predicting motivation to transfer. In addition, it points out that several factors work simultaneously to influence motivation to transfer EUC training.

Supervisory support in the pre‐ and post‐training environment is extremely crucial in determining EUC training success. Specifically, supervisors should be able to communicate to employees the purpose and importance of training, the relevance of computer training to their jobs and the outcomes expected.

This paper contributes to the literature by emphasizing the importance of supervisory support and individual characteristics in predicting motivation to transfer.

Three steps must be implemented if a training programme is to be successful. The first is the identification of needs to identify what training is required. The second is an analysis of the firm to identify the issues that will affect the ability of the firm to exploit new skills. The third is an evaluation of the training to ensure that sufficient resources are applied to implement and to integrate the training programme. These latter two steps come under the heading of learning transfer. The article presents the findings of an application of this approach. The analysis shows the richness of the information that results from this approach and outlines its operational importance for managers engaged in decision‐making or in the design of training programmes. In addition, it suggests the next steps in the research towards improving the tools available for the evaluation of training.

The effectiveness of the construction industry highly depends on the quality of the work practices, education and training. Based on the industry’s need to strengthen the productivity and performance improvement scopes in the construction education/training practices, this study aims to develop a guiding tool for designing new training programmes in different qualification levels considering the industry’s near-future challenges.

The study methodology encompassed literature reviews, experts’ discussions/reviews and problem-based communication approaches with qualitative methods to obtain a set of expected competency outcomes (COs) for each qualification level. The necessary mapping frameworks were used to display the cross section of each CO.

The study has presented a guiding model comprising three categories of qualification levels, where each consists of 12 COs with the mapping outcomes against learning domains of a conceptual framework. The guiding model also displays the detailed steps to develop detailed curriculums for different qualification levels of training.

Though the scope of the study was limited to the Sri Lankan context, the findings can be beneficial to many other countries in similar scenarios.

Overall study outcomes are expected to make an impact on the industry’s reskilling and upskilling practices resulting in a considerable level of improvement in the quality and productivity of work operations.

The experts’ reviews highlight the developed model as a valuable tool that provides a platform for upgrading the construction education and training practices based on the industry’s near-future circumstances in productivity and performance improvement practices, also revealing its future influences in training accreditation processes.

The purpose of this research is to expand cross‐cultural research and validate the Learning Transfer System Inventory in Ukraine. The researchers seek to translate the LTSI into Ukrainian and investigate the internal structure of this translated version of the questionnaire.

The LTSI is translated into Ukrainian using a rigorous translation methodology. The instrument was distributed to 511 (430 – usable data) individuals from various organizations throughout Ukraine. The exploratory factor analysis (common factor analysis with oblique rotation) was used to analyze the survey data.

The factor structure of the Ukrainian version of the LTSI (ULTSI) paralleled that of the original instrument. Most items loaded on appropriate factors with loadings over 0.4. Two factors (Opportunity to Use Learning and Performance‐Outcomes Expectations) did not emerge as clearly as in the original instrument and require further research.

The instrument was distributed to a convenience sample which limits the external validity of the study. Some translation issues may have possibly influenced low loadings for few items. The questionnaires distributed in the Eastern part of Ukraine were problematic because the population there is mostly Russian‐speaking while the instruments were in Ukrainian. The study provides evidence of construct validity of the LTSI in the Ukrainian business environment. With the limitations outlined above further research can address translation issues and language barrier problems. Some items may be included that will suit the Ukrainian environment better, which may resolve problems in two factors mentioned above. New methods of instrument translation could be utilized in future as well as alternative statistical analyses (Confirmatory Factor Analysis).

The ULTSI is now one of the first and few HRD tools in Ukraine. It can be used to assess the transfer climate in Ukrainian organizations in order to maximize the positive outcomes of the investment in training. As a diagnostic tool the ULTSI can provide necessary information about the environmental and organizational forces that are at play in a given organization, and it can help improve the results of training interventions.

This cross‐cultural study is one of the first efforts in existence to bridge the Ukrainian and Western cultures in terms of HRD. It provides a possibility for HRD methodology and theory to be introduced in Ukrainian businesses. At the same time, the study provides evidence of the construct validity and sound structure of the LTSI.

This paper's aim is to evaluate understanding and knowledge of the Adult Support and Protection (Scotland) Act 2007 in a sample of community nurses working in learning disability services in Scotland.

Ten community nurses who worked in learning disability services in one NHS area were tested at two time points, four months apart using a questionnaire designed for this study by researchers and practitioners. Level of previous national training in the Adult Support and Protection Act and length of time working with people with learning disabilities were recorded. Three domains of adult protection were included in the questionnaire: Principles of the Act and definitions; Adults at risk of harm; Protection, assessment, removal and banning orders.

Questionnaire scores varied widely overall and across the three domains. There was no correlation between individual scores and training or length of work experience. The level of knowledge was below what might have been expected for this group, given the level of training and experience. Carefully designed verification of the impact of nationally approved adult support and protection training is needed.

There is an absence of research in evaluating the impact of the approved Scottish Government training materials on staff knowledge and understanding of the 2007 Act, with staff attendance being taken as the main measure of training compliance. This was a small scale pilot study and recommendations are made for the scope and methods of evaluation.

Purpose: In recent times, ‘artificial intelligence (AI)’ has been pervasive even in organisations or at home. AI is defined as programming computers or other technological devices to act, react, respond, or assist the same way humans do. AI has undeniably made people’s lives easier. In organisations, the impact of AI is even more visible. The main aim of this chapter is to examine the significant role of future work skill’s (FWS) each component in the field of on-growing automation. The focus will be especially on emotional and social intelligence (ESI) (a key component of FWS) while adopting AI.

Need of the Study: In terms of human resource management (HRM), AI is useful for people management, payroll services, staff monitoring and improving the recruiting network, among other things. Even managers put their organisation’s job openings on the web and get applicant resumes electronically. People and employees in the organisation have become more advanced and innovative due to AI. A device obtains employee attendance, and human resource (HR) can track their employees and their organisation’s workforce data. HR has now been awarded more authority to manage and fix their employee’s problems because of AI. In a rapidly changing world, AI is affecting all aspects. AI is yearning to automate all of the jobs.

Methodology: Now a question arises how we can stay relevant in AI economic development? As humans, we learned that every issue is a problem of optimisation because we simply require human skills to develop, create and innovate new things. Therefore, researchers recognised that adopting sustainable growth skills encourages people to continue learning throughout their lives. Moreover, AI has enabled machines the ability to learn over time. Still, they will never be able to develop new ideas like human intelligence. A machine can use only one fixed data algorithm. Now humans have made significant progress in various fields with the help of FWS; without integrated computer sciences, brain science would not make such an outstanding achievement. On the other hand, human minds are masters of their intelligence, such as creativity, complex problem-solving, cognitive thinking, ESI and communication. Breakthrough human mind are masters of algorithms represented people have to understand new trends of technology around us, and the best way to move forward is to be aware, adapt and update skills.

Practical Implications: However, AI is required because, regardless of technological advancements, AI is leading Industry 4.0. The industry’s transformation is in 4.0, and hopefully, 5.0 will jump on board soon. Undoubtedly, AI should streamline the process and eliminate redundancy or administrative tasks.

Finding: AI can be more effective in organisations if they incorporate other FWS, particularly the soft human ESI skills, whereas AI is present everywhere, we can still not neglect FWS, especially ESI. So, this chapter highlights the important role of soft skills, that is, ESI and FWS, while adapting AI for an effective HRM.