Safety Implications of Variability in Autonomous Driving Assist Alerting

Cummings, Mary L.; Bauchwitz, Ben

doi:10.1109/tits.2021.3109555

Cited by 17 publications

(10 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A key component to understanding driver readiness is hand activity, as a distracted driver often has their hands off the wheel or on other devices like a phone. Rangesh et al (Rangesh et al, 2021) and Deo & Trivedi (Deo & Trivedi, 2019) show that driver hand activity is the most important component of models for prediction of driver readiness and takeover time, two metrics critical to safe control transitions in autonomous vehicles (Greer et al, 2023) (Cummings & Bauchwitz, 2021). Such driver-monitoring models take hand activity classes and held-object classes as input, among other components, as illustrated in Figure 1.…”

Section: Safety and Advanced Driver Assistance Systemsmentioning

confidence: 99%

Multi-View Ensemble Learning With Missing Data: Computational Framework and Evaluations using Novel Data from the Safe Autonomous Driving Domain

Greer¹,

Lulua²,

Gopalkrishnan³

et al. 2023

Preprint

View full text Add to dashboard Cite

Real-world applications with multiple sensors observing an event are expected to make continuously-available predictions, even in cases where information may be intermittently missing. We explore methods in ensemble learning and sensor fusion to make use of redundancy and information shared between four camera views, applied to the task of hand activity classification for autonomous driving. In particular, we show that a late-fusion approach between parallel convolutional neural networks can outperform even the best-placed single camera model. To enable this approach, we propose a scheme for handling missing information, and then provide comparative analysis of this late-fusion approach to additional methods such as weighted majority voting and model combination schemes.

show abstract

Section: Safety and Advanced Driver Assistance Systemsmentioning

confidence: 99%

Multi-View Ensemble Learning With Missing Data: Computational Framework and Evaluations using Novel Data from the Safe Autonomous Driving Domain

Greer¹,

Lulua²,

Gopalkrishnan³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…However, the deployment of CNNs presents serious safety challenges as there is a significant lack of human interpretability or intuition for the functioning of these Neural Network (NN) [15], even in the case of visual imagery. This is compounded by the variance in perception suite performance, where it simultaneously out-and under-preforms human capability leading to a cognition gap [16,17] making it difficult for drivers/operators to predict the vehicle's (mis-) behaviour with respect to unseen or unforeseen scenarios and edge-cases [18]. Where CNNs are used for object detection and classification, e.g.…”

Section: Capabilities and Limitations Of Camera-based Perception Systemsmentioning

confidence: 99%

Safety Assessment for Autonomous Systems' Perception Capabilities

Molloy¹,

McDermid²

2022

Preprint

View full text Add to dashboard Cite

Autonomous Systems (AS) are increasingly proposed, or used, in Safety Critical (SC) applications. Many such systems make use of sophisticated sensor suites and processing to provide scene understanding which informs the AS' decision-making. The sensor processing typically makes use of Machine Learning (ML) and has to work in challenging environments, further the ML-algorithms have known limitations,e.g., the possibility of false-negatives or false-positives in object classification. The well-established safety-analysis methods developed for conventional SC systems are not well-matched to AS, ML, or the sensing systems used by AS. This paper proposes an adaptation of well-established safety-analysis methods to address the specifics of perception-systems for AS, including addressing environmental effects and the potential failure-modes of ML, and provides a rationale for choosing particular sets of guidewords, or prompts, for safety-analysis. It goes on to show how the results of the analysis can be used to inform the design and verification of the AS and illustrates the new method by presenting a partial analysis of a road vehicle. Illustrations in the paper are primarily based on optical sensing, however the paper discusses the applicability of the method to other sensing modalities and its role in a wider safety process addressing the overall capabilities of AS.

show abstract

“…ADAS systems rely on computer vision-based lane tracking systems to maintain lateral control [3], and may combine computer vision with other sensing modalities to maintain space from other vehicles and detect obstacles [4][5][6]. However, recent research demonstrates that ADAS systems perform inconsistently in lateral and longitudinal control, including obstacle detection, warning and mitigations [7][8][9].…”

Section: Introductionmentioning

confidence: 99%

Unreliable Pedestrian Detection and Driver Alerting in Intelligent Vehicles

Cummings,

Bauchwitz

2024

IEEE Trans. Intell. Veh.

View full text Add to dashboard Cite

Vehicles with advanced driving assist systems that automatically steer, accelerate and brake are popular, but associated with increased driver distraction. This distraction, coupled with unreliable autonomous system performance, leads to vehicles that may be at higher risk for striking pedestrians. To this end, this study tested three consumer vehicles in two different model classes in a pedestrian crossing scenario. In 120 trials, one model never detected the pedestrian, nor alerted the driver. In 123 trials, the other model vehicles almost always detected the pedestrian, but in 35% of trials, alerted the driver too late. These cars were not consistent internally or with one another in pedestrian detections and responses, and only sparingly sounded any warnings. These intelligent vehicles also detected the pedestrian earlier if there were no established lane lines, suggesting that in well-marked areas, typically the case in for established crossings, pedestrians may be at increased risk of a possible conflict. This research demonstrates that artificial intelligence can lead to unreliable vehicle behaviors and warnings in pedestrian detection, potentially catching drivers off guard. These results further indicate industry needs to do more testing of intelligent systems, regulators should reevaluate the selfcertification approval process, and that more fundamental work is needed in academia around the performance and quality of technologies with embedded neural networks.

show abstract

Safety Implications of Variability in Autonomous Driving Assist Alerting

Cited by 17 publications

References 15 publications

Multi-View Ensemble Learning With Missing Data: Computational Framework and Evaluations using Novel Data from the Safe Autonomous Driving Domain

Multi-View Ensemble Learning With Missing Data: Computational Framework and Evaluations using Novel Data from the Safe Autonomous Driving Domain

Safety Assessment for Autonomous Systems' Perception Capabilities

Unreliable Pedestrian Detection and Driver Alerting in Intelligent Vehicles

Contact Info

Product

Resources

About