A scene comprehensive safety evaluation method based on binocular camera

Zhang, Xinyu; Shao, Wenbo; Zhou, Mo; Tan, Qifan; Li, Jun

doi:10.1016/j.robot.2020.103503

Cited by 15 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…from a lead vehicle/object and is applicable in CAVs such as adaptive cruise control (ACC) or automatic emergency braking (AEB). The safe distances formulated in [11], [12] considers ideal situations with reaction time by ego and lead vehicle and the required braking distance with the difference that [12] impose a safety margin. The ideal safe distance results in a long gap to be maintained to lead vehicles is inspired by Rule 126 of the UK Highway Code [15] but is not a realistic value applied by drivers on daily driving tasks.…”

Section: Consmentioning

confidence: 99%

“…The proposed formulation for longitudinal direction shall overcome the unrealistic distances defined in the literature. Therefore we target the acting distance identified as braking distance in Equation 1, and tune deceleration using both normal driving ranges as in [11], [13], and maximum feasible deceleration by a passenger car, similar to acting distance component in [12]. Therefore, the Equation is formulated as follows:…”

Section: A Proposed Metricsmentioning

confidence: 99%

See 1 more Smart Citation

Detecting Hazardous Events: A Framework for Automated Vehicle Safety Systems

Xiao

Geiger

Yatbaz

et al. 2022

2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

The driving domain is inherently dangerous. To develop connected and automated vehicles that can detect potential sources of harm, we must clearly define these hazardous events and metrics to detect them. The majority of driving scenarios we face do not materialise harm, but we often face potentially hazardous near-miss scenarios. Potential harm is difficult to quantify when harm is not materialised; thus, few metrics detect these scenarios in the absence of collision and even fewer datasets label non-collision-based hazardous events. This study focuses on detecting near-miss scenarios due to other actors since human error is the primary source of harm. We first provide a concise overview of current event-specific metrics. We then propose an event-agnostic detection framework that exploits vehicle kinematics to detect evasive manoeuvres early and dynamically calculate minimum safe distances. Given inconsistent dataset labelling methods and collision-focused events, we provide a preliminary study to demonstrate an eventagnostic and configurable dataset annotation technique to label hazardous events, even when harm is not materialised. We show promising results detecting hazardous scenes on a labelled simulation benchmark, GTACrash.

show abstract

Section: Consmentioning

confidence: 99%

Section: A Proposed Metricsmentioning

confidence: 99%

Detecting Hazardous Events: A Framework for Automated Vehicle Safety Systems

Xiao

Geiger

Yatbaz

et al. 2022

2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

show abstract

“…The author claims that the method is an original work and it avoids complicated vehicle searching and recognition in the video. Zhang et al (2020b) proposed a scene safety evaluation method based on a binocular camera. The model presents the uncertainty of the scene with a single value, which can be expressed as a quantitative evaluation of the safety of intended functionality (SOTIF).…”

Section: Implementing Sensors To Perceive the Environmentmentioning

confidence: 99%

A review of vehicle active safety control methods: From antilock brakes to semiautonomy

Ahangarnejad

Radmehr

Ahmadian

2020

Journal of Vibration and Control

View full text Add to dashboard Cite

A comprehensive review of technologies and approaches for active safety systems designed to reduce ground vehicle crashes, as well as the associated severity of injuries and fatalities, is provided. Active safety systems are commonly referred to as systems that can forewarn a driver of a potential safety hazard, or automatically intervene to reduce the likelihood of an accident without requiring driver intervention. The data from naturalistic drivers has shown that such systems are instrumental in improving vehicle safety in various conditions, particularly at higher speeds and under adverse road conditions. The increased integration of sensors, electronics, and real-time processing capabilities has served as one of the critical enabling elements in the widespread integration of active safety systems in modern vehicles. The emphasis is placed on control approaches for active safety systems and their progression over the years from antilock brakes to more advanced technologies that have nearly enabled semiautonomous driving. A review of key active safety control approaches for antilock braking, yaw stability, traction control, roll stability, and various collision avoidance systems is provided.

show abstract

“…With the development of computer vision and image processing, cameras have been the main perception devices. [1][2] Compared to other sensors, binocular camera provide richer information about the surrounding environment. Furthermore, cameras are cheap, have a small size, and consume little power [3][4].…”

Section: Introductionmentioning

confidence: 99%