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.

The growing availability of naturalistic driving datasets (NDDs) presents a valuable opportunity to develop various models for autonomous driving. However, while current NDDs include data on vehicles with and without intended driving behavior changes, they do not explicitly demonstrate a type of data on vehicles that intend to change their driving behavior but do not execute the behaviors because of safety, efficiency, or other factors. This missing data is essential for autonomous driving decisions. This study aims to extract the driving data with implicit intentions to support the development of decision-making models.

According to Bayesian inference, drivers who have the same intended changes likely share similar influencing factors and states. Building on this principle, this study proposes an approach to extract data on vehicles that intended to execute specific behaviors but failed to do so. This is achieved by computing driving similarities between the candidate vehicles and benchmark vehicles with incorporation of the standard similarity metrics, which takes into account information on the surrounding vehicles' location topology and individual vehicle motion states. By doing so, the method enables a more comprehensive analysis of driving behavior and intention.

The proposed method is verified on the Next Generation SIMulation dataset (NGSim), which confirms its ability to reveal similarities between vehicles executing similar behaviors during the decision-making process in nature. The approach is also validated using simulated data, achieving an accuracy of 96.3 per cent in recognizing vehicles with specific driving behavior intentions that are not executed.

This study provides an innovative approach to extract driving data with implicit intentions and offers strong support to develop data-driven decision-making models for autonomous driving. With the support of this approach, the development of autonomous vehicles can capture more real driving experience from human drivers moving towards a safer and more efficient future.

This research examines what motivates professional truck drivers to engage in eco-driving by linking their self-reports with objective driving scores.

Theory of Planned Behavior (TPB) is illustrated in an embedded, single-case study of a Finnish carrier with 17 of its truck drivers. Data are obtained through in-depth interviews with drivers, their fuel-efficiency scores generated by fleet telematics and a focus group session with the management.

Discrepancies between drivers’ intentions and eco-driving behaviors are illustrated in a two-by-two matrix that classifies drivers into four categories: ideal eco-drivers, wildcards, wannabes and non-eco-drivers. Attitudes, subjective norms and perceived behavioral control are examined for drivers within each category, revealing that drivers’ perceptions did not always align with the reality of their driving.

This study strengthens the utility of TPB through data triangulation while also revealing the theory’s inherent limitations in elucidating the underlying causes of its three antecedents and their impact on the variance in driving behaviors.

Managerial insights are offered to fleet managers and eco-driving solution providers to stipulate the right conditions for drivers to enhance fuel-efficiency outcomes of transport fleets.

This is one of the first studies to give a voice to professional truck drivers about their daily eco-driving practice.

The purpose of this study was to provide an outline of the process of developing an on-road driving test route and rating form. Comprehensive evaluation of medical fitness to drive should comprise of an off-road and an on-road assessment. Much research attention has focussed on the off-road phase of assessment, while there is less standardisation evident in the completion and measurement of the on-road phase of fitness-to-drive assessment.

A scholarship of practice approach was used to inform the development of an on-road test route and an associated generic on-road assessment tool that was guided by research evidence and best practice recommendations.

A step-by-step guide, outlining seven recommended phases in the development of an on-road route for the assessment of fitness to drive that aligns with best practice recommendations, was developed. A preliminary generic on-road assessment tool (the Maynooth–Trinity Driving Test) that includes higher-order cognition alongside element of strategic, tactical and operational driving ability was developed and piloted alongside the newly developed on-road test route.

This paper offers an overview of an approach to developing evidence-based on-road test routes and an associated generic assessment tool that may assist occupational therapists and on-road driving assessors establish a standard practice for testing on-road behaviour as part of a comprehensive approach to evaluate fitness to drive.

This study aims to improve the maneuverability and stability of four-wheel chassis in a small paddy field; a front axle swing steering four-wheel chassis with optimal steering is designed.

When turning, the front inner wheel stops and the rear inner wheel is in the following state. The hydraulic drive system of the walking wheel adopts a driving mode in which two front-wheel motors are connected in series and two rear wheel motors in parallel. The chassis uses a combination of a gasoline engine with a water cooling system, a CVT continuously variable transmission and a hydraulic drive system to increase the control capability. The front axle rotary chassis adopts a step-less variable speed engine and a hydraulic control system to solve the hydraulic stability of the chassis in uphill and downhill conditions so as to effectively control the over-speed of the wheel-side drive motors. Through the quadratic orthogonal rotation combination design test, the mathematical models of uphill and downhill front-wheel pressures and test factors are established.

The results show that the chassis stability is optimal when the back pressure is 0.5 MPa, and the rotating slope is 4°. The uphill and downhill pressures of the front wheels are 2.38 MPa and 1.5 MPa, respectively.

The influence of external changes on the pressure of hydraulic motors is studied through experiments, which lays the foundation for further research.

This paper aims to propose a dynamic parameter identification method based on sensitivity analysis for the 5-degree of freedom (DOF) hybrid robots, to solve the problems of too many identification parameters, complex model, difficult convergence of optimization algorithms and easy-to-fall into a locally optimal solution, and improve the efficiency and accuracy of dynamic parameter identification.

First, the dynamic parameter identification model of the 5-DOF hybrid robot was established based on the principle of virtual work. Then, the sensitivity of the parameters to be identified is analyzed by Sobol’s sensitivity method and verified by simulation. Finally, an identification strategy based on sensitivity analysis was designed, experiments were carried out on the real robot and the results were verified.

Compared with the traditional full-parameter identification method, the dynamic parameter identification method based on sensitivity analysis proposed in this paper converges faster when optimized using the genetic algorithm, and the identified dynamic model has higher prediction accuracy for joint drive forces and torques than the full-parameter identification models.

This work analyzes the sensitivity of the parameters to be identified in the dynamic parameter identification model for the first time. Then a parameter identification method is proposed based on the results of the sensitivity analysis, which can effectively reduce the parameters to be identified, simplify the identification model, accelerate the convergence of the optimization algorithm and improve the prediction accuracy of the identified model for the joint driving forces and torques.

Given the increasing importance of engineering insurance, it is still unclear which specific factors can enhance the role of engineering insurance as a risk transfer tool. This study aims to propose a hybrid approach to identify and analyze the key determinants influencing the consumption of engineering insurance in mainland China.

The empirical analysis utilizes provincial data from mainland China from 2008 to 2019. The research framework is a novel amalgamation of the generalized method of moments (GMM) model, the quantile regression (QR) technique and the random forest (RF) algorithm. This innovative hybrid approach provides a comprehensive exploration of the driving factors while also allowing for an examination across different quantiles of insurance consumption.

The study identifies several driving factors that significantly impact engineering insurance consumption. Income, financial development, inflation, price, risk aversion, market structure and the social security system have a positive and significant influence on engineering insurance consumption. However, urbanization exhibits a negative and significant effect on the consumption of engineering insurance. QR techniques reveal variations in the effects of these driving factors across different levels of engineering insurance consumption.

This study extends the research on insurance consumption to the domain of the engineering business, making theoretical and practical contributions. The findings enrich the knowledge of insurance consumption by identifying the driving factors specific to engineering insurance for the first time. The research framework provides a novel and useful tool for examining the determinants of insurance consumption. Furthermore, the study offers insights into the engineering insurance market and its implications for policymakers and market participants.

This study investigated the concerns and plans of construction professionals about building information modeling (BIM) implementation, found the acceptable BIM implementation driving forces and strategies for them and developed a prescriptive BIM implementation model to help understand how BIM implementation concerns, intentions, driving forces and strategies are connected.

This study employs a positivist paradigm with a hypothetico-deductive research strategy as well as concern-based adoption theory as a conceptual lens to distinguish construction professionals (CPs)' BIM implementation concerns and intentions. This implies that the forces driving BIM implementation intentions and concerns are related to BIM implementation methods and that their concentrations are proportional to the intensity of BIM implementation strategies. A 16-item questionnaire tailored to the operations of CPs was used for data collection. The data collected from respondents were utilized to evaluate the proposed model using structural equation modeling (SEM) techniques.

Findings from the data collected from the respondents revealed that CPs are concerned about the impact of BIM deployment on their time and service quality. Their main purpose was to take drives to learn more about BIM in order to pique their curiosity. Embracing the latest digital technology and beginning self-initiated BIM training are two strategies that would be quite effective in boosting BIM deployment.

The study identifies promising directions for future BIM implementation research and development. The study's findings imply that more theoretically motivated research, rather than just empirical research, is required to refine BIM implementation concerns.

The study has implications for the professional development of CPs as well as understanding the process of implementing BIM change. The study's findings will help to understand the resource system for assessing CPs' needs and concerns and selecting personalized BIM implementation strategies.

Before this study, BIM-related studies had ignored the concerns and goals of the CPs when it came to implementing BIM. Using the CPs' concerns and hopes for BIM implementation, a systemic BIM implementation model was developed that would help and speed up BIM adoption.