Do Autonomous Vehicle Driving Styles Affect User State?: A Preliminary Investigation

Abstract A bursting number of achievements in the autonomous vehicle industry have been obtained during the past decades. Various systems have been developed to make automated driving possible. Due to the algorithm used in the autonomous vehicle system, the performance of the vehicle differs from one to another. However, very few studies have given insight into the influence caused by implementing different algorithms from a human factors point of view. Two systems based on two algorithms with different characteristics are utilized to generate the two driving styles of the autonomous vehicle, which are implemented into a driving simulator in order to create the autonomous driving experience. User’s skin conductance (SC) data, which enables the evaluation of user’s cognitive workload and mental stress were recorded and analyzed. Subjective measures were applied by filling out Swedish occupational fatigue inventory (SOFI-20) to get a user self-reporting perspective view of their behavior changes along with the experiments. The results showed that human’s states were affected by the driving styles of different autonomous systems, especially in the period of speed variation. By analyzing users’ self-assessment data, a correlation was observed between the user “Sleepiness” and the driving style of the autonomous vehicle. These results would be meaningful for the future development of the autonomous vehicle systems, in terms of balancing the performance of the vehicle and user’s experience.

Download Full-text

Influence of Driving Experience on Distraction Engagement in Automated Vehicles

Transportation Research Record Journal of the Transportation Research Board ◽

10.1177/0361198119843476 ◽

2019 ◽

Vol 2673 (9) ◽

pp. 142-151 ◽

Cited By ~ 6

Author(s):

Dengbo He ◽

Birsen Donmez

Keyword(s):

Secondary Task ◽

Driving Simulator ◽

Phase 1 ◽

Task Engagement ◽

Automated Vehicles ◽

Automated Driving ◽

Novice Drivers ◽

Driving Experience ◽

Driving Skill ◽

The Difference

State-of-the-art vehicle automation requires drivers to visually monitor the driving environment and the automation (through interfaces and vehicle’s actions) and intervene when necessary. However, as evidenced by recent automated vehicle crashes and laboratory studies, drivers are not always able to step in when the automation fails. Research points to the increase in distraction or secondary-task engagement in the presence of automation as a potential reason. However, previous research on secondary-task engagement in automated vehicles mainly focused on experienced drivers. This issue may be amplified for novice drivers with less driving skill. In this paper, we compared secondary-task engagement behaviors of novice and experienced drivers both in manual (non-automated) and automated driving settings in a driving simulator. A self-paced visual-manual secondary task presented on an in-vehicle display was utilized. Phase 1 of the study included 32 drivers (16 novice) who drove the simulator manually. In Phase 2, another set of 32 drivers (16 novice) drove with SAE-level-2 automation. In manual driving, there were no differences between novice and experienced drivers’ rate of manual interactions with the secondary task (i.e., taps on the display). However, with automation, novice drivers had a higher manual interaction rate with the task than experienced drivers. Further, experienced drivers had shorter average glance durations toward the task than novice drivers in general, but the difference was larger with automation compared with manual driving. It appears that with automation, experienced drivers are more conservative in their secondary-task engagement behaviors compared with novice drivers.

Download Full-text

Analysis of Preference for Autonomous Driving Under Different Traffic Conditions Using a Driving Simulator

Journal of Robotics and Mechatronics ◽

10.20965/jrm.2015.p0660 ◽

2015 ◽

Vol 27 (6) ◽

pp. 660-670 ◽

Cited By ~ 6

Author(s):

Udara Eshan Manawadu ◽

◽

Masaaki Ishikawa ◽

Mitsuhiro Kamezaki ◽

Shigeki Sugano ◽

...

Keyword(s):

Autonomous Vehicles ◽

Driving Simulator ◽

Autonomous Driving ◽

Driving Experience ◽

The Road ◽

Traffic Conditions ◽

Passenger Vehicles ◽

New Type ◽

On The Road ◽

Near Future

<div class=""abs_img""><img src=""[disp_template_path]/JRM/abst-image/00270006/08.jpg"" width=""300"" /> Driving simulator</div>Intelligent passenger vehicles with autonomous capabilities will be commonplace on our roads in the near future. These vehicles will reshape the existing relationship between the driver and vehicle. Therefore, to create a new type of rewarding relationship, it is important to analyze when drivers prefer autonomous vehicles to manually-driven (conventional) vehicles. This paper documents a driving simulator-based study conducted to identify the preferences and individual driving experiences of novice and experienced drivers of autonomous and conventional vehicles under different traffic and road conditions. We first developed a simplified driving simulator that could connect to different driver-vehicle interfaces (DVI). We then created virtual environments consisting of scenarios and events that drivers encounter in real-world driving, and we implemented fully autonomous driving. We then conducted experiments to clarify how the autonomous driving experience differed for the two groups. The results showed that experienced drivers opt for conventional driving overall, mainly due to the flexibility and driving pleasure it offers, while novices tend to prefer autonomous driving due to its inherent ease and safety. A further analysis indicated that drivers preferred to use both autonomous and conventional driving methods interchangeably, depending on the road and traffic conditions.

Download Full-text

Get Ready for Being Chauffeured

Human Factors The Journal of the Human Factors and Ergonomics Society ◽

10.1177/0018720819872893 ◽

2019 ◽

Vol 62 (8) ◽

pp. 1322-1338 ◽

Cited By ~ 4

Author(s):

Kristin Mühl ◽

Christoph Strauch ◽

Christoph Grabmaier ◽

Susanne Reithinger ◽

Anke Huckauf ◽

...

Keyword(s):

Field Study ◽

Skin Conductance ◽

Driving Simulator ◽

Autonomous Vehicle ◽

Close Relation ◽

Autonomous Driving ◽

Autonomous Agent ◽

Driving Style ◽

Passenger Seat

Objective We investigated passenger’s trust and preferences using subjective, qualitative, and psychophysiological measures while being driven either by human or automation in a field study and a driving simulator experiment. Background The passenger’s perspective has largely been neglected in autonomous driving research, although the change of roles from an active driver to a passive passenger is incontrovertible. Investigations of passenger’s appraisals on self-driving vehicles often seem convoluted with active manual driving experiences instead of comparisons with being driven by humans. Method We conducted an exploratory field study using an autonomous research vehicle ( N = 11) and a follow-up experimental driving simulation ( N = 24). Participants were driven on the same course by a human and an autonomous agent sitting on a passenger seat. Skin conductance, trust, and qualitative characteristics of the perceived driving situation were assessed. In addition, the effect of driving style (defensive vs. sporty) was evaluated in the simulator. Results Both investigations revealed a close relation between subjective trust ratings and skin conductance, with increased trust and by trend reduced arousal for human compared with automation in control. Even though driving behavior was equivalent in the simulator when being driven by human and automation, passengers most preferred and trusted the human-defensive driver. Conclusion Individual preferences for driving style and human or autonomous vehicle control influence trust and subjective driving characterizations. Application The findings are applicable in human-automation research, reminding to not neglect subjective attributions and psychophysiological reactions as a result of ascribed control duties in relation to specific execution characteristics.

Download Full-text

Human-vehicle dynamic model with driver’s neuromuscular characteristic for shared control of autonomous vehicle

Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering ◽

10.1177/0954407020977108 ◽

2020 ◽

pp. 095440702097710

Author(s):

Wei Hanbing ◽

Wu Yanhong ◽

Chen Xing ◽

Xu Jin ◽

Rahul Sharma

Keyword(s):

Driving Simulator ◽

Unscented Kalman Filter ◽

Autonomous Vehicle ◽

Autonomous Driving ◽

Least Square ◽

Shared Control ◽

Driving Safety ◽

Dynamic State ◽

Commercial Application ◽

Steering Torque

Over a long period of time, the fully autonomous vehicle is far from commercial application. The concept of ‘human-vehicle shared control (HVSC)’ provides a promising solution to enhance autonomous driving safety. In order to characterize the evolution of the driver’s feature in the process of HVSC, a dynamics model of HVSC with the driver’s neuromuscular characteristic is proposed in this paper. It takes into account the driver’s neuromuscular characteristics, such as stretch reflection, feedback stiffness, etc. By designing a model predictive control (MPC) controller, the feedback of the vehicle’s state and steering torque is constructed. For validation of the model, driving simulation has been conducted in our table-based driving simulator. The vehicle state and the surface electromyography of the driver’s arm working muscle group are collected simultaneously. Subsequently, the hierarchical least square (HLS) parameter identification and unscented Kalman filter (UKF) observer is used to identify and estimate the important characteristic parameters respectively based on the experimental results. The comparisons show that the HVSC can characterize the vehicle’s dynamic state and the driver’s personalized characteristic can be identified by HLS. This paper will serve as a theoretical basis of control strategy allocation between the human and vehicle during shared control for L3 class autonomous vehicle.

Download Full-text

Function-Specific Uncertainty Communication in Automated Driving

International Journal of Mobile Human Computer Interaction ◽

10.4018/ijmhci.2019040105 ◽

2019 ◽

Vol 11 (2) ◽

pp. 75-97

Author(s):

Alexander Kunze ◽

Stephen J. Summerskill ◽

Russell Marshall ◽

Ashleigh J. Filtness

Keyword(s):

Augmented Reality ◽

Situation Awareness ◽

Driving Simulator ◽

Automated Driving ◽

Driving Experience ◽

Visual Variables ◽

Simulator Study ◽

The Impact

Conveying the overall uncertainties of automated driving systems was shown to improve trust calibration and situation awareness, resulting in safer takeovers. However, the impact of presenting the uncertainties of multiple system functions has yet to be investigated. Further, existing research lacks recommendations for visualizing uncertainties in a driving context. The first study outlined in this publication investigated the implications of conveying function-specific uncertainties. The results of the driving simulator study indicate that the effects on takeover performance depends on driving experience, with less experienced drivers benefitting most. Interview responses revealed that workload increments are a major inhibitor of these benefits. Based on these findings, the second study explored the suitability of 11 visual variables for an augmented reality-based uncertainty display. The results show that particularly hue and animation-based variables are appropriate for conveying uncertainty changes. The findings inform the design of all displays that show content varying in urgency.

Download Full-text

The Effect of Augmented Reality Cues on Glance Behavior and Driver-Initiated Takeover on SAE Level 2 Automated-Driving

Proceedings of the Human Factors and Ergonomics Society Annual Meeting ◽

10.1177/1071181321651004 ◽

2021 ◽

Vol 65 (1) ◽

pp. 1342-1346

Author(s):

Nayara de Oliveira Faria ◽

Coleman Merenda ◽

Richard Greatbatch ◽

Kyle Tanous ◽

Chihiro Suga ◽

...

Keyword(s):

Augmented Reality ◽

Situation Awareness ◽

Driver Assistance System ◽

User Study ◽

Driving Simulator ◽

Autonomous Vehicle ◽

Driver Assistance ◽

Automated Driving ◽

The Relationship ◽

Level 2

In the present paper, we present a user study with an advanced-driver assistance system (ADAS) using augmented reality (AR) cues to highlight pedestrians and vehicles when approaching intersections of varying complexity. Our major goal is to understand the relationship between the presence and absence of AR, driver-initiated takeover rates and glance behavior when using a SAE Level 2 autonomous vehicle. Therefore, a user-study with eight participants on a medium-fidelity driving simulator was carried out. Overall, we found that AR cues can provide promising means to increase the system transparency, drivers’ situation awareness and trust in the system. Yet, we suggest that the dynamic glance allocation of attention during partially automated vehicles is still challenging for researchers as we still have much to understand and explore when AR cues become a distractor instead of an attention guider.

Download Full-text

Function-Specific Uncertainty Communication in Automated Driving

10.4018/978-1-6684-3694-3.ch049 ◽

2022 ◽

pp. 1002-1026

Author(s):

Alexander Kunze ◽

Stephen J. Summerskill ◽

Russell Marshall ◽

Ashleigh J. Filtness

Keyword(s):

Augmented Reality ◽

Situation Awareness ◽

Driving Simulator ◽

Automated Driving ◽

Driving Experience ◽

Visual Variables ◽

Simulator Study ◽

The Impact

Download Full-text

Autonomous Vehicle Driving via Deep Deterministic Policy Gradient

Volume 3: 21st International Conference on Advanced Vehicle Technologies; 16th International Conference on Design Education ◽

10.1115/detc2019-97884 ◽

2019 ◽

Author(s):

Wenhui Huang ◽

Francesco Braghin ◽

Stefano Arrigoni

Keyword(s):

Driving Simulator ◽

Autonomous Vehicle ◽

Autonomous Driving ◽

Machine Learning Algorithms ◽

Gradient Algorithm ◽

Vehicle Model ◽

Human In The Loop ◽

Policy Gradient ◽

Discrete Action ◽

Continuous Actions

Abstract Autonomous driving has became one of the most hot trends in artificial intelligence area in recent years thanks to the machine learning algorithms. However, most of the autonomous driving studies are still limited to discrete action space. In this study, we propose to implement Deep Deterministic Policy Gradient algorithm for learning driving behavior over the continuous actions. For this purpose, a driving simulator is employed which interfaces with IPG CarMker software where the virtual environment and dynamical vehicle model can be built. “Human-in-the-loop” is performed in order to gather the data and a neural network which is implemented in Behavior Layer is trained to recognize two different scenarios-forward driving and stop. Based on the scenario the agent is dealing with, the actions are learnt and suggested from the DDPG algorithm. The experimental results show that DDPG algorithm is able to learn the optimal policy with continuous actions reliably for both scenarios.

Download Full-text

Intention-Aware Autonomous Driving Decision-Making in an Uncontrolled Intersection

Mathematical Problems in Engineering ◽

10.1155/2016/1025349 ◽

2016 ◽

Vol 2016 ◽

pp. 1-15 ◽

Cited By ~ 21

Author(s):

Weilong Song ◽

Guangming Xiong ◽

Huiyan Chen

Keyword(s):

Decision Making ◽

Autonomous Vehicles ◽

Driving Simulator ◽

Autonomous Vehicle ◽

Transition Process ◽

Autonomous Driving ◽

Driving Safety ◽

Time Step ◽

Reward Function ◽

Pass Through

Autonomous vehicles need to perform social accepted behaviors in complex urban scenarios including human-driven vehicles with uncertain intentions. This leads to many difficult decision-making problems, such as deciding a lane change maneuver and generating policies to pass through intersections. In this paper, we propose an intention-aware decision-making algorithm to solve this challenging problem in an uncontrolled intersection scenario. In order to consider uncertain intentions, we first develop a continuous hidden Markov model to predict both the high-level motion intention (e.g., turn right, turn left, and go straight) and the low level interaction intentions (e.g., yield status for related vehicles). Then a partially observable Markov decision process (POMDP) is built to model the general decision-making framework. Due to the difficulty in solving POMDP, we use proper assumptions and approximations to simplify this problem. A human-like policy generation mechanism is used to generate the possible candidates. Human-driven vehicles’ future motion model is proposed to be applied in state transition process and the intention is updated during each prediction time step. The reward function, which considers the driving safety, traffic laws, time efficiency, and so forth, is designed to calculate the optimal policy. Finally, our method is evaluated in simulation with PreScan software and a driving simulator. The experiments show that our method could lead autonomous vehicle to pass through uncontrolled intersections safely and efficiently.

Download Full-text

$$360^{\circ }$$ real-time and power-efficient 3D DAMOT for autonomous driving applications

Multimedia Tools and Applications ◽

10.1007/s11042-021-11624-2 ◽

2022 ◽

Author(s):

Carlos Gómez-Huélamo ◽

Javier Del Egido ◽

Luis Miguel Bergasa ◽

Rafael Barea ◽

Elena López-Guillén ◽

...

Keyword(s):

Embedded System ◽

Object Detection ◽

Real Time ◽

Driving Simulator ◽

Autonomous Driving ◽

Intelligent Vehicles ◽

Evaluation Tool ◽

Automated Driving ◽

Safe Driving ◽

Power Efficient

AbstractAutonomous Driving (AD) promises an efficient, comfortable and safe driving experience. Nevertheless, fatalities involving vehicles equipped with Automated Driving Systems (ADSs) are on the rise, especially those related to the perception module of the vehicle. This paper presents a real-time and power-efficient 3D Multi-Object Detection and Tracking (DAMOT) method proposed for Intelligent Vehicles (IV) applications, allowing the vehicle to track $$360^{\circ }$$ 360 ∘ surrounding objects as a preliminary stage to perform trajectory forecasting to prevent collisions and anticipate the ego-vehicle to future traffic scenarios. First, we present our DAMOT pipeline based on Fast Encoders for object detection and a combination of a 3D Kalman Filter and Hungarian Algorithm, used for state estimation and data association respectively. We extend our previous work ellaborating a preliminary version of sensor fusion based DAMOT, merging the extracted features by a Convolutional Neural Network (CNN) using camera information for long-term re-identification and obstacles retrieved by the 3D object detector. Both pipelines exploit the concepts of lightweight Linux containers using the Docker approach to provide the system with isolation, flexibility and portability, and standard communication in robotics using the Robot Operating System (ROS). Second, both pipelines are validated using the recently proposed KITTI-3DMOT evaluation tool that demonstrates the full strength of 3D localization and tracking of a MOT system. Finally, the most efficient architecture is validated in some interesting traffic scenarios implemented in the CARLA (Car Learning to Act) open-source driving simulator and in our real-world autonomous electric car using the NVIDIA AGX Xavier, an AI embedded system for autonomous machines, studying its performance in a controlled but realistic urban environment with real-time execution (results).

Download Full-text