Enhancing Autonomous Vehicle Decision-Making at Intersections in Mixed-Autonomy Traffic: A Comparative Study Using an Explainable Classifier

The transition to fully autonomous roadways will include a long period of mixed-autonomy traffic. Mixed-autonomy roadways pose a challenge for autonomous vehicles (AVs) which use conservative driving behaviours to safely negotiate complex scenarios. This can lead to congestion and collisions with human drivers who are accustomed to more confident driving styles. In this work, an explainable multi-variate time series classifier, Time Series Forest (TSF), is compared to two state-of-the-art models in a priority-taking classification task. Responses to left-turning hazards at signalized and stop-sign-controlled intersections were collected using a full-vehicle driving simulator. The dataset was comprised of a combination of AV sensor-collected and V2V (vehicle-to-vehicle) transmitted features. Each scenario forced participants to either take ("go") or yield ("no go") priority at the intersection. TSF performed comparably for both the signalized and sign-controlled datasets, although all classifiers performed better on the signalized dataset. The inclusion of V2V data led to a slight increase in accuracy for all models and a substantial increase in the true positive rate of the stop-sign-controlled models. Additionally, incorporating the V2V data resulted in fewer chosen features, thereby decreasing the model complexity while maintaining accuracy. Including the selected features in an AV planning model is hypothesized to reduce the need for conservative AV driving behaviour without increasing the risk of collision.

Effects of multimodal explanations for autonomous driving on driving performance, cognitive load, expertise, confidence, and trust

Advances in autonomous driving provide an opportunity for AI-assisted driving instruction that directly addresses the critical need for human driving improvement. How should an AI instructor convey information to promote learning? In a pre-post experiment (n = 41), we tested the impact of an AI Coach's explanatory communications modeled after performance driving expert instructions. Participants were divided into four (4) groups to assess two (2) dimensions of the AI coach's explanations: information type ('what' and 'why'-type explanations) and presentation modality (auditory and visual). We compare how different explanatory techniques impact driving performance, cognitive load, confidence, expertise, and trust via observational learning. Through interview, we delineate participant learning processes. Results show AI coaching can effectively teach performance driving skills to novices. We find the type and modality of information influences performance outcomes. Differences in how successfully participants learned are attributed to how information directs attention, mitigates uncertainty, and influences overload experienced by participants. Results suggest efficient, modality-appropriate explanations should be opted for when designing effective HMI communications that can instruct without overwhelming. Further, results support the need to align communications with human learning and cognitive processes. We provide eight design implications for future autonomous vehicle HMI and AI coach design.

Incorporating Situation Awareness Cues in Virtual Reality for Users in Dynamic in-Vehicle Environments

The increasing ubiquity and mobility of virtual reality (VR) devices has introduced novel use cases, one of which is using VR in vehicles, both human-driven and fully automated. However, the effects of the adoption of VR-in-the-car on user task performance, safety, trust, and perceived risk are still largely unknown or not fully understood. Blocking out the physical world and substituting it with a virtual environment has many potential benefits including fewer distractions and greater productivity. However, one shortcoming of this seclusion is losing situation awareness which becomes critical in dynamic, in-vehicle environments, even when the user is not in the driver's seat. Hence, this study aims to understand the effects of providing VR users with situation awareness cues about the real world, when riding in a human-driven or a fully automated car. The results of this driving simulator experiment provide valuable insights into passengers' experience and their information needs while immersed in VR environments. Identifying passengers' unique challenges and needs, as well as developing solutions for them, is expected to improve users' travel experience towards a wider adoption of VR devices.

When Self-Driving Fails: Evaluating Social Media Posts Regarding Problems and Misconceptions about Tesla’s FSD Mode

With the recent deployment of the latest generation of Tesla’s Full Self-Driving (FSD) mode, consumers are using semi-autonomous vehicles in both highway and residential driving for the first time. As a result, drivers are facing complex and unanticipated situations with an unproven technology, which is a central challenge for cooperative cognition. One way to support cooperative cognition in such situations is to inform and educate the user about potential limitations. Because these limitations are not always easily discovered, users have turned to the internet and social media to document their experiences, seek answers to questions they have, provide advice on features to others, and assist other drivers with less FSD experience. In this paper, we explore a novel approach to supporting cooperative cognition: Using social media posts can help characterize the limitations of the automation in order to get information about the limitations of the system and explanations and workarounds for how to deal with these limitations. Ultimately, our goal is to determine the kinds of problems being reported via social media that might be useful in helping users anticipate and develop a better mental model of an AI system that they rely on. To do so, we examine a corpus of social media posts about FSD problems to identify (1) the typical problems reported, (2) the kinds of explanations or answers provided by users, and (3) the feasibility of using such user-generated information to provide training and assistance for new drivers. The results reveal a number of limitations of the FSD system (e.g., lane-keeping and phantom braking) that may be anticipated by drivers, enabling them to predict and avoid the problems, thus allowing better mental models of the system and supporting cooperative cognition of the human-AI system in more situations.

Take-Over Requests after Waking in Autonomous Vehicles

Autonomous vehicles (AVs) enable drivers to devote their primary attention to non-driving-related tasks (NDRTs). Consequently, AVs must provide intelligibility services appropriate to drivers’ in-situ states and in-car activities to ensure driver safety, and accounting for the type of NDRT being performed can result in higher intelligibility. We discovered that sleeping is drivers’ most preferred NDRT, and this could also result in a critical scenario when a take-over request (TOR) occurs. In this study, we designed TOR situations where drivers are woken from sleep in a high-fidelity AV simulator with motion systems, aiming to examine how drivers react to a TOR provided with our experimental conditions. We investigated how driving performance, perceived task workload, AV acceptance, and physiological responses in a TOR vary according to two factors: (1) feedforward timings and (2) presentation modalities. The results showed that when awakened by a TOR alert delivered >10 s prior to an event, drivers were more focused on the driving context and were unlikely to be influenced by TOR modality, whereas TOR alerts delivered <5 s prior needed a visual accompaniment to quickly inform drivers of on-road situations. This study furthers understanding of how a driver’s cognitive and physical demands interact with TOR situations at the moment of waking from sleep and designs effective interventions for intelligibility services to best comply with safety and driver experience in AVs.

Trust in Autonomous Cars: Exploring the Role of Shared Moral Values, Reasoning, and Emotion in Safety-Critical Decisions

OBJECTIVE

Autonomous cars (ACs) controlled by artificial intelligence are expected to play a significant role in transportation in the near future. This study investigated determinants of trust in ACs.

BACKGROUND

Trust in ACs influences different variables, including the intention to adopt AC technology. Several studies on risk perception have verified that shared value determines trust in risk managers. Previous research has confirmed the effect of value similarity on trust in artificial intelligence. We focused on moral beliefs, specifically utilitarianism (belief in promoting a greater good) and deontology (belief in condemning deliberate harm), and tested the effects of shared moral beliefs on trust in ACs.

METHOD

We conducted three experiments (N = 128, 71, and 196, for each), adopting a thought experiment similar to the well-known trolley problem. We manipulated shared moral beliefs (shared vs. unshared) and driver (AC vs. human), providing participants with different moral dilemma scenarios. Trust in ACs was measured through a questionnaire.

RESULTS

The results of Experiment 1 showed that shared utilitarian belief strongly influenced trust in ACs. In Experiment 2 and Experiment 3, however, we did not find statistical evidence that shared deontological belief had an effect on trust in ACs.

CONCLUSION

The results of the three experiments suggest that the effect of shared moral beliefs on trust varies depending on the values that ACs share with humans.

APPLICATION

To promote AC implementation, policymakers and developers need to understand which values are shared between ACs and humans to enhance trust in ACs.

Drivers trust, acceptance, and takeover behaviors in fully automated vehicles: Effects of automated driving styles and driver's driving styles

Automated Vehicle (AV) technology has the potential to significantly improve driver safety. Unfortunately, drivers could be reluctant to ride with AVs due to their lack of trust and acceptance of AVs' driving styles. The present study investigated the effects of the designed driving style of AV (aggressive/defensive) and driver's driving style (aggressive/defensive) on driver's trust, acceptance, and take-over behavior in a fully AV. Thirty-two participants were classified into two groups based on their driving styles using the Aggressive Driving Scale and experienced twelve driving scenarios in either an aggressive AV or a defensive AV. Results revealed that driver's trust, acceptance, and takeover frequency were significantly influenced by the interaction effects between AV's driving style and driver's driving style. General estimating equations were conducted to analyze the relationships between driver's trust, acceptance, and take over frequency. The results showed that the effect of driver's trust in AVs on takeover frequency was mediated by driver's acceptance of AVs. These findings implied that driver's trust and acceptance of AVs could be enhanced when the designed AV's driving style aligned with driver's own driving style, which in turn, reduce undesired take over behavior. However, the "aggressive" AV driving style should be designed carefully considering driver safety.

Drivers’ Age and Automated Vehicle Explanations

Automated vehicles (AV) have the potential to benefit our society. Providing explanations is one approach to facilitating AV trust by decreasing uncertainty about automated decision-making. However, it is not clear whether explanations are equally beneficial for drivers across age groups in terms of trust and anxiety. To examine this, we conducted a mixed-design experiment with 40 participants divided into three age groups (i.e., younger, middle-age, and older). Participants were presented with: (1) no explanation, or (2) explanation given before or (3) after the AV took action, or (4) explanation along with a request for permission to take action. Results highlight both commonalities and differences between age groups. These results have important implications in designing AV explanations and promoting trust.

Exploring Personalised Autonomous Vehicles to Influence User Trust

Trust is a major determinant of acceptance of an autonomous vehicle (AV), and a lack of appropriate trust could prevent drivers and society in general from taking advantage of such technology. This paper makes a new attempt to explore the effects of personalised AVs as a novel approach to the cognitive underpinnings of drivers’ trust in AVs. The personalised AV system is able to identify the driving behaviours of users and thus adapt the driving style of the AV accordingly. A prototype of a personalised AV was designed and evaluated in a lab-based experimental study of 36 human drivers, which investigated the impact of the personalised AV on user trust when compared with manual human driving and non-personalised AVs. The findings show that a personalised AV appears to be significantly more reliable through accepting and understanding each driver’s behaviour, which could thereby increase a user’s willingness to trust the system. Furthermore, a personalised AV brings a sense of familiarity by making the system more recognisable and easier for users to estimate the quality of the automated system. Personalisation parameters were also explored and discussed to support the design of AV systems to be more socially acceptable and trustworthy.