自动驾驶汽车与行人交互中的沟通界面设计：基于行人过街决策模型的评估

自动驾驶汽车要进入人车混行的普通道路, 需确保与过街行人之间的交互安全和效率。为解决这一问题, 高等级自动驾驶汽车往往在车辆外部装置显示设备, 即外部人机界面(eHMIs)以和行人沟通信息。在具体设计上, 已有研究主要采用文字、图形、投影等视觉沟通形式, 传达车辆状态(是否在自动驾驶模式)、意图和对行人的过街建议等沟通信息, 并在真实路段实验、虚拟场景及实验室实验等情境中评估了界面的使用对行人过街意向、速度和准确性等指标的影响。然而, 以行人为中心的外部界面设计需系统地支持行人过街决策前各阶段的信息加工需求。因此, 我们结合行人过街决策过程和情境意识理论, 提出行人与自动驾驶汽车交互中的动态过街决策模型, 从行人认知加工视角评估各种界面的沟通效果。评估的结果启示, eHMIs应促进行人对车辆信息的感知、理解和预测。在感知阶段, 应采用多种类型界面、多呈现载体相结合, 增强信息的可识别性。在理解阶段, 需结合文字说明、合理选择沟通视角、信号标准化和培训提高可理解性。在预测阶段, 应结合车辆内隐运动信息, 帮助行人快速准确获取车辆未来行动意图。更重要的是, 未来研究应关注在多行人、多车辆混行情境下的信息沟通设计及其对行人的影响。理论方面, 未来研究也需要关注外部界面如何通过自下而上的通路影响情境意识和心智模型的形成。     

Factors of acceptability,acceptance and usage for non-rail autonomous public transport vehicles: A systematic literature review

Non-rail autonomous public transport vehicles have emerged over the last few years. Technical progress in automation has resulted in a growing number of autonomous shuttle pilot experiments. Although these systems are technologically feasible, determining the extent to which they correspond to users’ needs and expectations remains a major issue. In order to answer that question, we conducted a systematic review which synthesizes the literature regarding the acceptability and willingness to use this type of autonomous public transport. This literature review allowed us to identify 39 documents addressing 70 factors of acceptability, acceptance and usage of non-rail autonomous public transport vehicles. The most cited factors in the literature concern service characteristics (times, schedules, fares) and safety issues (road-safety, on-board security). Factors related to automation level, comfort and access to the vehicle feature appear to a lesser extent. Acceptance is also related to personal factors, such as socio-demographics, travel habits, and personality. This review could be of interest to designers and manufacturers of non-rail autonomous public transport vehicles, as well as policy makers, and assist with the successful implementation of autonomous public transport services which are better adapted and meet the needs of all potential users.     

The impact of a dedicated lane for connected and automated vehicles on the behaviour of drivers of manual vehicles

Connected and automated vehicles (CAVs) are expected to enhance traffic efficiency by driving at shorter time headways, and traffic safety by shorter reaction times. However, one of the main concerns regarding their deployment is the mixed traffic situation, in which CAVs and manually driven vehicles (MVs) share the same road.This study investigates the behavioural adaptation of MV drivers in car-following and lane changing behaviour when they drive next to a dedicated lane (DL) for CAVs and compares that to a mixed traffic situation. The expectation is that in a mixed traffic situation, the behavioural adaptation of MV drivers is negligible due to lower exposure time and scarce platoons, while concentrating the CAVs on one dedicated lane may cause significant behavioural adaptation of MV drivers due to a higher exposure time and conspicuity of CAV platoons.Fifty-one participants were asked to drive an MV on a 3-lane motorway in three different traffic scenarios, in a fixed-base driving simulator: (1) Base, only MVs were present in traffic, (2) Mixed, platoons of 2–3 CAVs driving on any lane and mixed with MVs, (3) DL, platoons of 2–3 CAVs driving only on a DL. The DL was recognizable by road signs and a buffer demarcation which separated the DL from the other lanes. A moderate penetration rate of 43% was assumed for CAVs.During the drives, the car following headways and the accepted merging gaps by participants were collected and used for comparisons of driving behaviour in different scenarios.Based on the results, we conclude that there is no significant difference in the driving behaviour between Base and Mixed scenarios at tested penetration rate, confirming our research expectation. However, in DL scenario, MV drivers drove closer to their leaders specially when driving on the middle lane next to the platoons and accepted shorter gaps (up to 12.7% shorter at on-ramps) in lane changing manoeuvres. Dedicating a lane to CAVs increases the density of CAV platoons on one lane and consequently their conspicuity becomes higher. As a result, MV drivers are influenced by CAV platoons on a DL and imitate their behaviour.The literature suggests that dedicating a lane to CAVs improves the traffic efficiency by providing more possibilities for platooning. This study shows that implementing such a solution will affect the driving behaviour of human drivers. This should be taken into consideration when evaluating the impacts of dedicated lanes on traffic efficiency and traffic safety.     

Public acceptance of driver state monitoring for automated vehicles: Applying the UTAUT framework

Driver state monitoring (DSM) systems aim to measure driver/occupant state, considering factors such as fatigue, workload, attentiveness, and wellbeing. They are influential for some vehicles on the road today, but as we move towards higher levels of automation their use is expected to become even more important. Uncertainty around public perception of these systems is a potentially limiting factor, with potential concerns for privacy, data storage, perceived utility, and reliability amongst other things posing potential threats. A survey was conducted based on the Universal Theory of Acceptance and Use of Technology (UTAUT) framework, which collected 311 individual responses to explore the acceptability of this technology in specific relation to conditionally and partially automated vehicles. This paper identified that the factors of effort expectancy, performance expectancy, social influence, and attitudes towards using new technology are all positively related to overall behavioural intention, whereas anxiety is negatively related. Other demographic influences are discussed, and recommendations made for future research and marketing/implementation of DSM systems to ensure acceptance by future users.     

External human-machine interfaces: Effects of message perspective

Future automated vehicles may be equipped with external Human-Machine Interfaces (eHMIs). Currently, little is known about the effect of the perspective of the eHMI message on crossing decisions of pedestrians. We performed an experiment to examine the effects of images depicting eHMI messages of different perspectives (egocentric from the pedestrian’s point of view: WALK, DON’T WALK, allocentric: BRAKING, DRIVING, and ambiguous: GO, STOP) on participants’ (N = 103) crossing decisions, response times, and eye movements. Considering that crossing the road can be cognitively demanding, we added a memory task in two-thirds of the trials. The results showed that egocentric messages yielded higher subjective clarity ratings than the other messages as well as higher objective clarity scores (i.e., more uniform crossing decisions) and faster response times than the allocentric BRAKING and the ambiguous STOP. When participants were subjected to the memory task, pupil diameter increased, and crossing decisions were reached faster as compared to trials without memory task. Regarding the ambiguous messages, most participants crossed for the GO message and did not cross for the STOP message, which points towards an egocentric perspective taken by the participant. More lengthy text messages (e.g., DON’T WALK) yielded a higher number of saccades but did not cause slower response times. We conclude that pedestrians find egocentric eHMI messages clearer than allocentric ones, and take an egocentric perspective if the message is ambiguous. Our results may have important implications, as the consensus among eHMI researchers appears to be that egocentric text-based eHMIs should not be used in traffic.     

Measuring driver-vehicle cooperation: Development and validation of the Human-Machine-Interaction-Interdependence Questionnaire (HMII)

The development of partially and highly autonomous vehicles that take over parts of the driving task will result in changes in e.g. the responsibilities, interface and system design, task allocation and communication between driver and automated vehicle. To support this change and the increasing space of cooperative possibilities between driver and vehicle, generally accepted design principles and preconditions for successful driver-vehicle cooperation (Directability, Mutual Predictability, Joint Goals and Mutual Task-Dependency) were defined. However, research lacks validated scales measuring the proposed basic principles of driver-vehicle cooperation. Furthermore, a theory is missing that links those basic principles with a theory that enables an understanding of the influence of drivers' perception of the autonomous vehicle and context on driver-vehicle cooperation. Therefore, this work links the basic theoretical principles of driver-vehicle cooperation with the social psychological Theory of Interdependence and their dimensions (Conflict, Power, Mutual Dependence, Information Certainty, Future Interdependence). Filling the gap of missing validated scales for the principles of driver-vehicle cooperation, this study provides the development and validation of the Human-Machine-Interaction-Interdependence (HMII) questionnaire.In two studies, the new HMII questionnaire to measure drivers' perception in driver-vehicle cooperation was developed. In the first study (n = 94), items for the perception of the situation were transferred from the original Theory of Interdependence to the driver-vehicle cooperation. A 7-dimension model was identified via Explorative Factor Analysis. In the second study (n = 314), the model and items were validated via confirmatory factor analysis. A seven-factor model (Power, Conflict, Mutual Dependence Information Certainty: System to Human, Information Certainty (two levels) Future Interdependence (two levels) with 33 items showed a good fit to the data, chi² = 841, p < .001, adjusted chi² = 1.77, SRMR = 0.071. In sum, this questionnaire can help designers evaluate the impact of their system designs on driver perceptions of the driver-vehicle cooperation.     

The determinants of Portuguese preference for vehicle automation: A descriptive and explanatory study

During the last century, innovation of automated vehicles (AVs) technologies are successively maturing while progressively excluding the human intervention in vehicle driving. The objective of this paper was to analyze the determinants of Portuguese drivers’ decision to adopt AVs technologies, in an under explored context, where the driver of contemporary vehicles does all or part of the dynamic driving task (DDT) in comparison to vehicles equipped with Automated Driving systems (ADS) where the driver can become a passenger temporarily or permanently. In addition, willingness-to-pay for ADS estimates were also investigated. This study data was collected through a survey designed and deployed in Portugal. A mixed logit model was estimated, and the results obtained are in line with the literature of AVs in a number of determinants, but also highlights differences that can be explained by the Portuguese cultural, social and economic context. Overall, 83.7% of the Portuguese drivers favor contemporary vehicles, today, and among those who prefer vehicles with ADS, highly educated drivers’ are willing to pay, on average, 65,671 € for Conditional AVs, 31,185 € for Highly AVs, and about 28,622 € for Full AVs.     

Visualizing the autonomous vehicle’s maneuvers – Does an ecological interface help to increase the hedonic quality and safety?

This paper examines whether ecological speed information describing ongoing driving maneuvers during automated driving enhances the hedonic quality and driving safety immediately after a driving takeover. Visualizing maneuvers and trajectories has already proven effective. However, planned acceleration and deceleration in an automated vehicle have not yet been investigated. Therefore, this paper assesses how an automated vehicle’s speed control information might be presented by an ecological interface. Besides a possible increase in the hedonic quality, this information might enhance safe behavior of the human driver when it comes to a takeover. To assess these two aspects, 43 drivers participated in a dynamic driving simulator study. Using a within-subject design, two scenarios were used to compare an ecological interface, dynamically visualizing speed changes, to a conventional pop-up interface, using pop-up icons to visualize speed changes. The experimental results indicate that ecological feedback and conventional pop-up feedback do not differ regarding the hedonic quality, which was reflected by the state anxiety, usefulness, and satisfaction with the overall human-machine interface (HMI). Nonetheless, the post-hoc questionnaire on situational awareness showed a significantly lower rating for the ecological interface which may be the result of a more automatic and subconscious processing of the information given. Analyzing the takeover performance, the initial takeover time was comparably low for both interfaces. However, concerning safety, the ecological interface significantly enhanced the lateral control after takeover, and the drivers looked at the vehicle mirrors significantly earlier. In conclusion, the results show that the information given by the ecological interface may help drivers cope with a sudden takeover in a faster and more controlled way. Future applications of these findings might serve to enhance the acceptance and safety of semi-autonomous vehicles by implementing ecological interfaces.     

A projective assessment of autonomous motivation in children: Correlational and experimental evidence

This article presents two studies aimed at validating a new TAT-like projective measure of autonomous motivation in children. Study 1 assesses the validity of the new measure by correlating it with self-report questionnaires of autonomous motivation, positive and negative affect, task value and mastery goal orientation. Study 2 is an experiment in which autonomous motivation is manipulated and then assessed with the new projective measure and with a self-report scale. Results of both studies support the validity of the new projective measure. In study 2, regression analysis suggests that the new projective measure is sensitive to aspects of experimentally induced autonomous motivation that are not captured by a self-report measure.

Sensorimotor representation and knowledge-based reasoning for spatial exploration and localisation

We investigate a hybrid system for autonomous exploration and navigation, and implement it in a virtual mobile agent, which operates in virtual spatial environments. The system is based on several distinguishing properties. The representation is not map-like, but based on sensorimotor features, i.e. on combinations of sensory features and motor actions. The system has a hybrid architecture, which integrates a bottom-up processing of sensorimotor features with a top-down, knowledge-based reasoning strategy. This strategy selects the optimal motor action in each step according to the principle of maximum information gain. Two sensorimotor levels with different behavioural granularity are implemented, a macro-level, which controls the movements of the agent in space, and a micro-level, which controls its eye movements. At each level, the same type of hybrid architecture and the same principle of information gain are used for sensorimotor control. The localisation performance of the system is tested with large sets of virtual rooms containing different mixtures of unique and non-unique objects. The results demonstrate that the system efficiently performs those exploratory motor actions that yield a maximum amount of information about the current environment. Localisation is typically achieved within a few steps. Furthermore, the computational complexity of the underlying computations is limited, and the system is robust with respect to minor variations in the spatial environments.