Skip to article
AHFE AHFE Discoveries AHFE International Open Access Journal
AHFE OPEN ACCESS OPEN ACCESS CONFERENCE PROCEEDINGS PAPER

The Evaluation of AR-HUD Visual Augmentation Method Based on Situation Awareness

School of Mechanical Engineering, Southeast University, Nanjing, Jiangsu Province, 211189, China ↗
Publication year
Digital Object Identifier10.54941/ahfe1005854

Abstract

Augmented Reality Head-Up Display (AR-HUD) enhances driving safety and experience by overlaying virtual information into the driver’s forward view. However, AR-HUD may lead to attention capture, causing the driver to focus excessively on the AR-HUD and neglect other critical environmental information. Therefore, it is crucial to study how to balance the driver’s attention allocation between AR-HUD information and environmental information to ensure effective situational awareness (SA). The study based on Situational Awareness Theory, evaluates the impact of different AR-HUD augmentation methods (boxes, arrows, and shadows) on drivers' SA through analyzing eye movement data and Situation Awareness Rating Technique (SART) data. The results show that the different augmentation methods significantly affect drivers' SA levels. The box-type augmentation method performs the best overall and effectively improves the driver's SA. The study provides a theoretical foundation and practical guidance for the optimization and application of AR-HUD design.

INTRODUCTION

Augmented Reality Head-Up Display (AR-HUD) plays a significant role in enhancing driving safety and further optimizes the driving experience by providing real-time navigation and traffic information (Azuma et al., 2001). Drivers can access necessary information more directly, thereby reducing inconveniences caused by difficulties in obtaining information during driving (Zhou et al., 2024). AR-HUD can project directly critical information, such as navigation and warnings, into the driver’s field of view, reducing the need for glancing down at the dashboard and thereby mitigating the risk of traffic accidents caused by visual distractions (Cheng et al., 2021). Studies indicate that AR-HUD significantly improves drivers’ situational awareness (SA) and reaction time for road conditions. However, AR-HUD may lead drivers to overlook some information to increase the risk of accidents due to the characteristic of “attention capture”(Mackworth, 1965). AR-HUD may introduce new distraction issues, particularly impacting the driver’s visual attention. Research has shown that the overlap between

© 2025. Published by AHFE Open Access. All rights reserved. 494

Research on the Evaluation of AR-HUD Visual Augmentation Method 495

HUD content and unexpected stimuli in the real world may increase the frequency of inattentional blindness (Oh et al., 2016). The study showed that the effects of AR graphic placement and the relative position of AR graphics to unpredictable stimuli on inattentional blindness by measuring its occurrence rate and reaction times. The findings reveal that unpredictable stimuli unexpectedly overlapping with AR graphics significantly increase the likelihood of neglecting unforeseen events and result in longer reaction times (Chen et al., 2023).

The concept of SA was first introduced by Endsley, defined as “the ability to perceive, comprehend, and predict environmental elements within a specific time and space” (Endsley, 1995). The theory has undergone multiple stages of development, with its applications evolving from initial use in the military to broader domains such as aviation, healthcare, and cybersecurity (Stanton et al., 2017). Research has shown that the impact of situation awareness on driving performance, revealing that drivers with higher situation awareness are better able to accurately assess the behavior of autonomous vehicles in unexpected driving scenarios, thereby increasing their trust and acceptance of autonomous systems (Avetisyan et al., 2022). The study demonstrates that prolonged warning delays (6 seconds) result in significantly lower situation awareness in drivers compared to shorter pre- alert warnings (3 seconds). This decline is primarily attributed to drivers’ overreliance on such systems, making them less capable of responding promptly to hazards during system failures (Tan et al., 2022).

In summary, existing research on AR-HUD lacks sufficient focus on whether the design of AR-HUD information influences driver distraction. Therefore, this study integrates SA to explore the design of AR-HUD, aiming to achieve appropriate visual attention allocation for drivers and ultimately improve driving safety.

METHODOLOGY

Participants 15 participants (9 male, 6 female), aged from 18 to 35 (M = 26.67, SD = 3.958) with more 3 years driving experience were participated in the study. All participants had normal visual abilities, with uncorrected or corrected vision above 0.6, and were in good mental health. Before the experiment, all participants were required to maintain a regular routine, ensuring sufficient sleep and rest.

Experimental Equipment (1) Driving Simulator Hardware: The equipment includes a 4K-165Hz high-resolution display consisting of three screens manufactured by Samsung. Additionally, the system is equipped with an adjustable semi- bucket seat (XDracing), a steering wheel, and pedals (Logitech, G923 TRUEFORCE). The eye movement data of the driver was collected using the Dikablis Glasses 3, a glasses-style eye tracker.

(2) SCANeR Studio Driving Simulation Software: SCANeR Studio is a professional driving simulation software used to simulate road conditions, weather, time, and autonomous vehicle takeover scenarios. A desktop computer is placed to the right of the simulated driving seat, responsible for running the SCANeR Studio driving simulation program (see Figure 1).

Figure 1: Experimental equipment.
Original PDF Figure 1:

Experimental Materials (1) The simulated driving experimental environment was constructed using SCANeR Studio software, including two types of driving scenarios: speed limit and obstacle avoidance. Simulated driving videos were recorded and divided into 10-second segments for experimental presentation. The enhancement color was set to red (FF0000) with warning semantics, and the AR-HUD interface was designed using Adobe Illustrator, including three enhancement styles: box, arrow, and shadow (see Figure 2).

Figure 2: Enhancement styles.
Original PDF Figure 2:

Research on the Evaluation of AR-HUD Visual Augmentation Method 497

(2) SART Situation Awareness Rating Technique. Participants were required to complete the SART questionnaire after finishing the experiments to assess their SA under different scenarios. Subjective evaluations of the current scenario were conducted from three dimensions: demand, supply, and understanding (see Table 1). The questionnaire data were collected by the Wenjuanxing platform.

Table 1. SART scale.

Table available in the original PDF paper.

Phase Project Definition Rating Low to high 1∼7 Demand(D) Uncertainty Degree of unknown or ambiguous information in the current situation (1 = completely clear, 7 = very unclear)

Low to high 1∼7

Complexity Complexity of the situation or task (1 = very simple, 7 = extremely complex) Time Pressure Low to high 1∼7 Impact of time constraints on task completion (1 = no pressure, 7 = extremely urgent)

Low to high 1∼7

Supply(S) Information Quality

Accuracy, reliability, and relevance of the information obtained (1 = completely useless, 7 = very reliable) Information Quantity Low to high 1∼7 Sufficiency of the amount of information obtained (1 = completely insufficient, 7 = fully sufficient)

Low to high 1∼7

Familiarity with Information

Understanding and familiarity with the information (1 = completely unfamiliar, 7 = very familiar) Low to high 1∼7 Understanding(U) Clarity Clarity of understanding the current situation (1 = completely unclear, 7 = very clear)

Low to high 1∼7

Confidence Confidence in one’s decisions and judgments (1 = no confidence at all, 7 = very confident) Sense of Control Low to high 1∼7 Ability to control the situation and its changes (1 = no control at all, 7 = fully in control)

Low to high 1∼7

Adaptability Ability to adapt to changes in the situation (1 = completely unable to adapt, 7 = highly adaptable)

Experimental Process At first, participants were informed with the experimental procedure and content. After wearing the eye tracker and completing the calibration process, the experiment began with the random presentation of Task 1 or Task 2. The eye tracker recorded the participants’ the first fixation time in different areas. After completing the experiment, participants were required to fill out the SART questionnaire (see Figure 3).

Task 1: The vehicle initially drived at the speed of 70 km/h. Participants were required to adjust their speed promptly based on the speed limit

displayed on the AR-HUD. The eye tracker recorded the first fixation time on the speed limit indicator.

Task 2: Participants were required to avoid potential collisions with cyclists based on the enhanced information provided by the AR-HUD. The eye tracker recorded the first fixation time on the target stimulus.

Figure 3: Experimental process.
Original PDF Figure 3:

RESULTS

Eye Tracking Data Analysis Based on the descriptive statistics (see Table 2) and one-way ANOVA (see Table 3) of first fixation time for enhanced prompts in the two tasks, the analysis revealed the relationships between different enhancement methods and first fixation time under distinct driving tasks (see Figure 4). The results indicated the following:

In Task 1, the box and arrow, shadow enhancement methods had significant differences in first fixation time. The arrow and shadow methods had no significant difference. The box enhancement method had significantly longer first fixation time compared to the other two methods.

In Task 2, the arrow and box, shadow enhancement methods had significant difference in first fixation time. The box and shadow methods had no significant difference. The arrow enhancement method had significantly longer first fixation time than the other two methods.

Table 2. Descriptive statistical analysis of first fixation time.

Table available in the original PDF paper.

Task Enhancement Method

N Mean Standard Deviation

Standard Error

95% Confidence Interval of the Difference Lower Bound Upper Bound

Box 15 .42560 .060512 .015624 .39209 .45911

Speed limit Box 15 1.21153 .248162 .064075 1.07411 1.34896 Arrow 15 .50720 .087261 .022531 .45888 .55552 Shadow 15 .45940 .108599 .028040 .39926 .51954 Obstacle avoidance Arrow 15 .75940 .100521 .025954 .70373 .81507 Shadow 15 .46267 .048377 .012491 .43588 .48946

Research on the Evaluation of AR-HUD Visual Augmentation Method 499

Table 3. One-way ANOVA for first fixation time.

Table available in the original PDF paper.

Task Experimental Condition

95% Confidence Interval of the Difference Significance Lower Standard Error Bound Upper Bound

(I) Enhancement Method

(J) Enhancement Method

Mean Difference (I-J)

Box Arrow .704333* .059997 <0.001 .58325 .82541 Speed limit

Box Arrow −.333800* .026755 <0.001 −.38779 −.27981

Shadow .752133* .059997 <0.001 .63105 .87321 Arrow Box −.704333* .059997 <0.001 −.82541 −.58325 Shadow .047800 .059997 .430 −.07328 .16888 Shadow Box −.752133* .059997 <0.001 −.87321 −.63105 Arrow −.047800 .059997 .430 −.16888 .07328 Obstacle avoidance Shadow −.037067 .026755 .173 −.09106 .01693 Arrow Box .333800* .026755 <0.001 .27981 .38779 Shadow .296733* .026755 <0.001 .24274 .35073 Shadow Box .037067 .026755 .173 −.01693 .09106 Box Arrow −.333800* .026755 <0.001 −.38779 −.27981

Figure 4: The first fixation time under different tasks.
Original PDF Figure 4:

SART Scores Analysis Based on the descriptive statistics (see Table 4) and one-way ANOVA (see Table 5) of SART scores for enhanced prompts in the two tasks, the analysis revealed the relationships between different enhancement methods and driving tasks (see Figure 5). The results indicated the following.

In Task 1 and Task 2, the arrow and box, shadow enhancement methods had significant differences in SART scores. The box and shadow methods had no significant difference. The SART scores for the arrow enhancement method were significantly lower than the other two methods.

Table 4. Descriptive statistical analysis of SART scores.

Table available in the original PDF paper.

Task Enhancement Method

N Mean Standard Deviation

Standard Error

95% Confidence Interval of the Difference Lower Bound Upper Bound

Box 15 25.26667 3.881580 1.002220 23.11712 27.41621

Speed limit Box 15 25.13333 2.748160 .709572 23.61145 26.65521 Arrow 15 13.93333 1.791514 .462567 12.94123 14.92544 Shadow 15 25.60000 2.947154 .760952 23.96792 27.23208 Obstacle avoidance Arrow 15 13.78571 3.042736 .813205 12.02889 15.54254 Shadow 15 24.00000 3.464102 .866025 22.15411 25.84589

Table 5. One-way ANOVA for SART scores.

Table available in the original PDF paper.

Task Experimental Condition 95% Confidence Interval of the Difference

Mean Difference (I-J)

Standard Error

Significance Lower Bound

Upper Bound

(I) Enhancement Method

(J) Enhancement Method Box Arrow 11.200000* .929698 <0.001 9.32379 13.07621

Speed limit

Box Arrow 11.480952* 1.296578 <0.001 8.86435 14.09755

Shadow −.466667 .929698 .618 −2.34287 1.40954 Arrow Box −11.200000* .929698 <0.001 −13.07621 −9.32379 Shadow −11.666667* .929698 <0.001 −13.54287 −9.79046 Shadow Box .037067 .026755 .173 −.01693 .09106 Arrow −.296733* .026755 <0.001 −.35073 −.24274 Obstacle avoidance Shadow 1.266667 1.253962 .318 −1.26393 3.79726 Arrow Box −11.480952* 1.296578 <0.001 −14.09755 −8.86435 Shadow −10.214286* 1.276867 <0.001 −12.79111 −7.63746 Shadow Box −1.266667 1.253962 .318 −3.79726 1.26393 Box Arrow 10.214286* 1.276867 <0.001 7.63746 12.79111

Figure 5: The SART scores under different tasks.
Original PDF Figure 5:

Research on the Evaluation of AR-HUD Visual Augmentation Method 501

CONCLUSION

In summary, this study conducted a simulated driving experiment to compare and evaluate the effects of different AR-HUD enhancement methods on drivers’ SA in two common driving scenarios: speed limit and obstacle avoidance.

The analysis of SART scores demonstrated that the box and shadow AR- HUD enhancement methods had a positive impact on drivers’ SA. However, the analysis of first fixation time revealed that, in the speed limit task, the box enhancement method strongly attracted drivers’visual attention, reducing their ability to adapt to changes in speed limit indicators, which could negatively impact driving safety. In contrast, the shadow enhancement method allowed drivers to quickly notice both road stimuli and speed limit changes in both scenarios, facilitating a more balanced allocation of visual attention across different areas. Therefore, the shadow enhancement method is identified as the optimal design approach. The findings can provide valuable insights for designing AR-HUD enhancement methods in vehicles. It can highlight the importance of ensuring that drivers maintain a high level of SA and achieve balanced visual attention allocation across different areas, thereby improving driving safety and mitigating the negative effects of attention capture caused by AR-HUD.

REFERENCES

Avetisyan L, Ayoub J and Zhou F (2022) ‘Investigating explanations in conditional and highly automated driving: The effects of situation awareness and modality’, Transportation research. Part F, Traffic psychology and behaviour, 89:456–466, doi: 10.1016/j.trf.2022.07.010. Azuma R, Baillot Y, Behringer R, Feiner S, Julier S and MacIntyre B (2001) Recent

advances in augmented reality, IEEE, Los Alamitos, doi: 10.1109/38.963459. Chen W, Song J, Wang Y, Wu C, Ma S, Wang D, Yang Z and Li H (2023) ‘Inattentional blindness to unexpected hazard in augmented reality head- up display assisted driving: The impact of the relative position between stimulus and augmented graph’, Traffic injury prevention, 24(4):344–351, doi: 10.1080/15389588.2023.2186735. Cheng, D., Wang, Q., Liu, Y., Chen, H., Ni, D., Wang, X., Yao, C., Hou,

Q., Hou, W., & Luo, G. (2021). Design and manufacture AR head-mounted displays: A review and outlook. Light: Advanced Manufacturing, 2(3), 350–369. doi: 10.37188/lam.2021.024. Endsley MR (1995) ‘Toward a Theory of Situation Awareness in Dynamic Systems’, Human factors, 37(1):32–64, doi: 10.1518/001872095779049543. Mackworth N H. Visual noise causestunnel vision[J]. Psychonomic science, 1965,

3(1-12): 67–68. Oh HJ, Ko SM and Ji YG (2016) ‘Effects of Superimposition of a Head- Up Display on Driving Performance and Glance Behavior in the Elderly’, International journal of human-computer interaction, 32(2):143–154, doi: 10.1080/10447318.2015.1104155. Stanton, N. A., Salmon, P. M., Walker, G. H., & Jenkins, D. P. (2017). Human

Factors Methods: A Practical Guide for Engineering and Design. CRC Press. doi: 10.1201/9781315587246.

Tan X, Zhang Y and Wang J (2022) ‘Assessing the potential impacts of connected vehicle systems on Driver’s situation awareness and driving performance’, Transportation research. Part F, Traffic psychology and behaviour, 84:177–193, doi: 10.1016/j.trf.2021.11.016. Zhou C, Qiao W, Hua J and Chen L (2024) ‘Automotive Augmented Reality Head-Up

Displays’, Micromachines (Basel), 15(4):442, doi: 10.3390/mi15040442.

Cover of the proceedings volume Intelligent Human Systems Integration (IHSI 2025): Integrating People and Intelligent Systems

Explore the complete digital volume

Intelligent Human Systems Integration (IHSI 2025): Integrating People and Intelligent Systems

Read this individual research paper free through the original publisher, or explore the full digital proceedings book.

ISBN 978-1-964867-36-6

Explore this ISBN in the AHFE Bookstore Listed in the July 2026 bookstore catalogue; search by ISBN to confirm current availability · Digital book purchase optional

Explore Related Research

Related research 8 matches

More papers from this volume

← Back to Research ExplorerBrowse the full library →