Human intervention is critical in situations where the autonomous system issues a takeover request in conditionally driving. Effective human-machine collaboration and safe transition require the timely and accurate identification of the driver's takeover intentions. Given the challenges associated with the opacity of decision-making processes in standalone end-to-end models for inferring takeover intentions, this paper introduces an approach for inferring takeover intention based on driver-hand trajectory estimation, which leverages the physical and observable actions as an interpretable indicator. We first presented a comprehensive framework designed to acquire preliminary data on the driver's hand and steering wheel position. This framework utilized MediaPipe for the acquisition of hand coordinates, EfficientDet for steering wheel localization, and MiDaS for depth estimation. Then, the Extended Kalman Filter (EKF) model was integrated into the Gated Recurrent Unit (GRU) network (GRU-EKF) to estimate hand movements and subsequently predict takeover intention combining current-time trajectory with estimated future states. The approach was validated through simulated experiments of human-machine driving interaction. In addition, a further examination is conducted to analyze the influence of six distinct combinations of historical and estimation horizons on model performance. Results show that the GRU-EKF methodology outperforms other models in terms of overall performance metrics. Using shorter estimation horizons can lead to higher accuracy and precision, whereas increasing the estimation horizon to 3 s notably improves the model's ability to correctly classify different takeover intentions. The findings can offer valuable perspectives for the design and enhancement of takeover process functionalities of Advanced Driver-Assistance Systems.

Advancements in autonomous driving technologies continue to revolutionize transportation, yet the full realization of self-driving vehicles remains hampered by several critical challenges. Automated vehicles continue to encounter significant challenges in perception, prediction, and decision-making, while their low-level modules are relatively mature and robust. Conversely, humans surpass machines in terms of high-level intelligence but may suffer from control performance degradation. To coalesce the strength of the human and machine, a novel collaboration scheme is proposed to compensate for the prediction-decision uncertainty via human guidance while providing low-level control feedback to the driver. This approach generates multiple decision candidates and corresponding predictions for other road users using a transformer-based socially compliant generative adversarial network (SCGAN). The driver can assist in choosing the appropriate candidate using the context-understanding capability, while concurrently, the control projection of this chosen decision guides the driver to achieve the desired objective via haptic steering feedback. The haptic feedback can reflect the decision uncertainties enabled by the decision-control projection of the intention estimation of the ego vehicle. A Type-II fuzzy controller is utilized to determine the control authority to account for the complexity of the future movement. We verify the effectiveness of the proposed algorithm through a real-time human-in-the-loop experiment, including an ablation study and comparisons with other human-machine collaboration schemes. The results demonstrate that the proposed scheme can minimize human-machine conflicts while increasing system safety.

New energy applications in transportation have emerged as a key strategy to reduce carbon emissions, but their performance in cold regions remains underexplored. We developed a stacking model to estimate energy consumption, evaluating the economic and environmental competitiveness of six diesel trucks (DTs) and their corresponding battery electric trucks (BETs) in cold regions. The results indicate that most BETs still exhibit a considerable cost advantage over DTs in low ambient temperatures. This advantage can be further enhanced under off-peak charging and favorable battery warranty conditions. From an environmental perspective, battery electric trucks produce higher CO2 and PM2.5 emissions but lower NOx than diesel trucks. These findings suggest that BETs have superior economic performance and potential to reduce emissions if supported by a cleaner grid mix. Our analysis provides new insights into the environmental and economic performance of BETs in cold regions, highlighting the conditions needed to improve their competitiveness.

Lane-changing behaviour is more complex and riskier than lane-keeping, posing challenges for autonomous vehicle decision-making. To address this issue, this study modelled the decision-making process of lane-changing behaviour by combining Stackelberg game theory with the potential field method. The potential field method analyzes dynamic interactions between multiple vehicles, while game theory quantifies the efficiency and safety benefits of lane-changing. Subsequently, a decision-making model for multi-vehicle interaction was established by integrating game theory and the potential field method. Finally, the lane-changing trajectory was optimised for maximum driving benefit. A simulation of lane-changing behaviour was used to validate the performance of the proposed model. The results indicated that the proposed method leads to a higher average velocity (20.07 m/s), faster lane changing (3.14 s) and lower computational costs (0.075 s) than the rule-based method (19.21 m/s, 4.04 and 0.17 s, respectively). This finding can provide a reference for reasonable decision-making for autonomous vehicles.

Conditional Automated Driving (CAD) has attracted widespread attention due to the substantial gap in achieving fully autonomous driving, wherein an essential endeavor entails determining the transition timing between automated and manual driving modes. Driver cognitive workload serves as a crucial indicator for identifying transition timing, while its precise determination is challenging with discrete workload levels in previous studies. To address this issue, this work develops a dual-stage learning framework to quantify driver cognitive workload continuously. Specifically, a semi-supervised co-training strategy is first designed to approximate workload values, and then supervised contrastive learning is employed to align them with their feature representations in the latent space. A novel driver workload dataset is constructed for the evaluation, and experimental results demonstrate that our proposed approach outperforms other state-of-the-art baselines in estimation accuracy. Furthermore, the rationality of quantified cognitive workload is analyzed through the driver’ subjective assessment, indicating it is a more reliable solution for achieving the driving authority transition.

In conditionally automated driving, traffic safety problems would occur if the driver does not properly take over the control authority when the request of automated system arises. Therefore, this study proposes XGBoost learning method considering risk potential field to predict the takeover quality in conditionally automated driving under different levels of cognitive non-driving related tasks (NDRTs). Thirty participants drive on two experimental conditions: manual driving is following an automated driving during which the driver is asked to perform NDRTs. Drivers’ physiological features of different cognitive states are exploited to model multi-level takeover quality prediction. This investigation also gives an insight into the main effects of the selected prediction variables on the takeover quality. The proposed model performance within different time windows is assessed using multiple evaluation metrics and compared with other methods. Results show that the prediction accuracy within the time windows of 7–10 s, 5–7 s, 3–5 s and 1–3 s is 0.87, 0.85, 0.85 and 0.90, respectively. The XGBoost model has the best performance of different time windows under each level of takeover quality compared to the other three machine learning models. Our findings can effectively predict the takeover quality and assist automated driving safely in interactions between drivers and automated vehicles.

Drivers in highly automated vehicles (AVs) are not required to continuously monitor the AV system. However, they must be prepared to take over vehicles when requested. Therefore, it is necessary to design an in-cabin interface that allows drivers to adapt their levels of preparedness to the likelihood of control transition. This study examined the interaction effects of (1) takeover request modality (TORM), (2) non-driving related tasks (NDRT), (3) takeover lead time (TOLT) on takeover performance. We conducted a driver-in-the-loop experiment involving 32 participants based on 15 takeover requests (TORs) for each NDRT. The Particle swarm optimization algorithm combined with multilayer perceptron learning was used for multi-objective balance optimization of five performance indicators. Results showed that the utilization of tactile-auditory prompts with 5 s and 9 s TOLTs exhibited positive performance in music listening and no task situations. The integration of tactile-auditory or tactile-visual cues with 7 s and 9 s TOLTs yielded favorable results in reading scenarios, whereas the tactile cues demonstrated efficacy in video watching scenarios. The situation of visual with 5 s TOLT reached the optimal balance of the five optimization objectives when the driver performed no task. This finding can offer valuable guidance to design interfaces in highly automated vehicles.

The lane-changing maneuvers of vehicles at freeway merging areas can result in a substantial increase in crash risk. The implementation of a connected and autonomous vehicle (CAV) environment is anticipated to decrease the associated conflict risk of such maneuvers, by utilization of surrounding traffic information. To investigate the potential conflict situations that may arise during multi-vehicle interactions by taking advantage of CAV environment, this study proposes a Bayesian hierarchical algorithm to identify traffic conflicts in real-time during mandatory drivers’ merging behavior. The Bayesian hierarchical model consists of observation vehicle conflict layer and prior distribution layer. The MCMC sampling method was utilized to calibrate the model parameters by generating a Markov chain. Two methods including image discrimination and posterior distribution comparison are combined to ensure reliable convergence diagnosis of Markov chain. The results show that the Markov chain for each parameter gradually stabilizes over iterations and the posterior probability density is consistent with the specified prior function and likelihood function. Additionally, the flexibility of this approach significantly enhances the ability to analyze traffic conflict from the real world using applied statistics. The findings of this research have the potential to encourage cautious driving behaviors and provide useful insights for driver safety in the future, especially in the context of CAVs.

Drivers will be required to maintain readiness to take over manual control of the automated vehicle occasionally during the highly automated driving until the advent of fully vehicle automation. Users can engage in cognitive tasks in this level of automated driving. However, it is unclear whether the engagement in cognitive secondary activities would impair or increase the driver's capacity to take over control authority. To explore such issue, four highly automated takeover scenarios were designed to study the individual and interaction effects of secondary tasks and time windows on takeover performance. Thirty participants engaged in similar cognitive tasks under different levels. A control group did not conduct any secondary tasks. Drivers had to resume vehicle control in four different scenarios while engaging in secondary tasks. The transition situation happened when a broken-down vehicle suddenly appears in front of the ego vehicle. Results show that there is significant difference in brake force, lateral offset and pitch, and no significant difference in sideslip angle-0 left, gas pedal, yaw speed and roll speed under different cognitive tasks. The takeover time under the condition of 0-back cognitive task is higher than that of no task (control group) and other two tasks, indicating that the proper workload of secondary task enhance alertness compared with the no-task, and high workload of secondary task could increase reaction time of takeover. The interaction effect between secondary task and time window on acceleration and brake force is not obvious. The significant difference exists in both individual and interaction effect of secondary task and time window on lateral offset and pitch. The findings could potentially provide reference for the design of safe and efficient highly automated driving systems during transition process.

The risky lane-changing manoeuvre of vehicles often occurs at expressway entrances, which would result in a high crash risk in the freeway system and significantly impact its safety. The highly anticipated environment of connected and autonomous vehicles (CAVs) is expected to reduce the associated crash risk of lane changing by offering various types of driving support, which utilise surrounding traffic information. The modelling crash risk under the environment of CAV driving during mandatory lane changing in merging areas faces new challenges due to the novelty of CAVs and subsequent shortage of data. To explore such risk situation of multi-vehicle interaction at expressway entrances, this study proposed a supervised learning algorithm and a Bayesian hierarchical model to assess risk levels and predict the probability of risk occurrence at different risk levels of interactive vehicles in real time of mandatory driving behaviour during the merging process. The learning algorithm, based on XGBoost, was exploited to classify risk levels. The Bayesian hierarchical model was used to analyse the probability of real-time risk comprising vehicle physical state layer, multi-vehicle interaction layer and risk probability layer. The probabilistic model parameters were calibrated using Markov Chain Monte Carlo (MCMC) Gibbs sampling method. The K-fold cross validation method was used to validate the proposed model of risk level. The probabilistic model validity was tested through posterior prediction of P-value. The quantitative risk estimation of CAVs through a few merging cases was conducted. Results show that the identification accuracy of slight, low, moderate and high risk is 94.24%, 85.82%, 84.16% and 79.69%, respectively. The P-value of Durbin-Watson's posterior, normal hypothesis, test distribution symmetry and kurtosis are all close to 0.5. Therefore, the method of real-time risk assessment is convergent and has good fitting. This research can promote cautious driving behaviours and provide reference for driver's decision making in the long term under the environment of CAVs.

To study the influence of the coupling relationship between primary and secondary tasks on driving safety, four driving experiments are designed, including Bluetooth calling, conversation, screen touch operation and in-vehicle radio operation. The vehicle operation data such as standard deviation of horizontal speed, standard deviation of longitudinal speed and steering entropy are collected. The driver visual data such as entropy rate of fixation area, standard deviation of horizontal viewing angle and vertical viewing angle, and average glance speed are collected. The coupling model of primary and secondary tasks in the vehicle is established to determine the coupling coordination degree between them. The driving proportion threshold of driver's secondary tasks and the risk level of various secondary tasks are obtained. The results show that the coupling degree of normal driving and Bluetooth communication can reach high quality coordination level with high safety. Similarly, the coordination level between normal driving and talking is also high. The coupling degree of normal driving and operating radio can reach medium coordination level with general safety. The coupling degree of normal driving and touch screen operation can only achieve low coordination, which is a relatively dangerous driving state.

To better construct the coordinated development of the comprehensive transportation system, the prediction model of comprehensive freight index in China was studied. Basic indicators of production and supply, benefits and macroeconomic were selected. Cluster analysis was used to divide the indicators into 13 leading indicators and 13 synchronous indicators. Factor analysis was used to analyze the importance of each leading indicator. The characteristic quantities of freight index were determined and the Granger causality test was conducted between the four indexes and the five characteristic quantities respectively. The relationship model between each freight index and characteristic quantity was constructed, and the validity of the model was verified. Results show that the freight indexes are volatile. There is a stable relationship between characteristic quantity and freight index. The error of forecast and actual value of is within 10%, verifying the validity of the model. The results can provide theoretical support for forecasting the development trends of freight transport among various transport modes. The comprehensive transport will be improved to provide a decision-making basis for the national macro-department to make policies.

Passenger transportation structure (PTS) features a huge effect on the proficiency of the territorial comprehensive traffic framework. Therefore, one of the important means to improve its efficiency is to rationalize and optimize the passenger transportation structure. From a static point of view, this paper puts forward user optimum thought to meet the demand of regional passenger optimal thought and constructed an optimization model of the passenger transportation structure on the basis of land and energy consumption. Through analysis of the passenger transportation systems of Harbin, Daqing, and Yichun in China, a model was designed by simulation and the proportion of the traveler transportation organization were determined. The method of fuzzy comprehensive was used to assess the traveler transportation arrangement, which is based on optimization model and real-world data. The effectiveness of the optimization model was demonstrated by analysis of the evaluation results. The displayed method and estimation results can help transportation organizers and policy-makers in deciding future traveler transportation foundation speculation needs.

为减少低能见度下无信号交叉口过街行人与车辆的交通事故,开展了考虑能见度影响的车辆与过街行人冲突识别研究.结合车辆行人相对位置、速度、加速度、车辆尺寸等信息,构建了过街行人与车辆冲突识别模型,确定了基于人-车间距的交通环境能见度测量方法,给出车辆速度与能见度之间的关系模型,在此基础上对模型进行修正,并验证了模型的有效性.结果表明:该冲突识别模型可对过街行人与车辆冲突进行有效识别,冲突识别的准确率为82.4％,该研究可为车-路协同下的无信号交叉口行人和车辆冲突识别提供决策,进而提高低能见度下行人与车辆的安全性.

为定量分析交通流中车辆间的交互影响,开展了主干路车辆交互速度变化模型研究,通过类比势能场理论中的引力和斥力,界定了势能场影响区域的概念,将目标车运行时与周围车辆的吸引和排斥作用归结为势能场影响区域交互面积的变化,提出考虑势能场影响区域的车辆交互分析方法,建立了目标车速度变化量与势能场影响区域交互面积的线性模型.采用P3-DT北斗高精度定位测向机采集车辆坐标和速度,标定模型参数.用该模型计算目标车速度变化量,并与实测数据进行对比.结果表明:模型计算值与实际值之间的误差小于15％,车道变换时间越短,目标车与目标车道后方车辆间的交互作用越明显,后车的减速操作越迅速,验证了模型的有效性.该模型将传统微观交通流分析中的车速与车辆间距两大因素归一为势能场影响区域交互面积,可为微观交通流中的多车交互研究提供方法,并为自动驾驶车辆提供速度控制策略.

Connected vehicle (CV)technologies offer promising solutions to several problems in transportation systems. The trajectory data generated from CV technology can be used to identify real-time conflicts in intersections. To perform such identification, accurate vehicle localisation should be obtained to clearly recognise the conflicts between left-turning vehicles and straight-through vehicles in the opposite direction at the signal control intersection. This study presents a CV framework that uses the two-way time of arrival to locate the vehicles on the basis of the Intelligent Vehicle Infrastructure Cooperative Environment. Kalman Filter (KF) is used to improve the accuracy of the vehicle location, and the corresponding algorithm is used to estimate the vehicle trajectory to obtain the vehicle kinematics information via the on-board system. The traffic conflict areas of the left-turning vehicles and straight-through vehicles in the opposite direction are determined through vehicle trajectory extrapolation, and the left-turn collision at the signal intersection is identified using the post-encroachment time algorithm and vehicle movement information. In addition, Anderson-Darling and modified Kolmogorov-Smirnov tests are performed to verify the goodness of fit of the data. Results show that the vehicle speed and localisation errors of the proposed method decreased by 66.67 % and 83.33 % compared with the results before filtering, respectively. Moreover, the results of the conflict recognition method based on CV trajectory reconstruction is consistent for both goodness of fit tests under real-time communication conditions. This study can provide driving decision for drivers of left-turning vehicles under the Intelligent Vehicle Infrastructure Cooperative Environment and provide technical support for the research and development of left-turn anti-collision systems.

The causal relationship between zonal freight turnover and gross domestic product (GDP) are receiving increasing attention to coordinate the balanced development of freight transportation and zonal economic level effectively. However, studies on the causality direction between freight transportation and economic have so far been in debate. To understand the relationship of causality direction between transportation and economic development for different zones which can provide decision support for public policies, econometrics is used to analyze the relationship between them. This paper investigates the relationship between GDP and freight turnover for economic zones of stationary series and non-stationary sequences using the vector autoregressive (VAR) and the vector error correction models (VECM). The impulse response analysis and variance decomposition are conducted to verify the effectiveness of the models. The Granger causality test is exploited to discover the relationship between transportation and economic development in each economic zone. The data on freight transportation and GDP in China from 2003 to 2018 is used. Results show that the relationship between freight turnover and GDP in the Northeast economic zone is bidirectional. A unidirectional relationship exists between freight turnover and GDP in the Circum Bohai-Sea, the Pearl River, Middle Part, Southwest, and Northwest zone. And the Granger causality is not obvious in the Yangtze River economic zone. In addition, suggestions for the zonal development of transportation and economic systems are provided. This study can provide a basis to adopt relevant policies and measures of sustainable development between transportation and economic growth for different zones.