Adaptive intervention logic for automated driving systems based on injury risk minimization

Vangi, Dario; Virga, Antonio; Gulino, Michelangelo-Santo

doi:10.1177/0954407020931228

Cited by 7 publications

(8 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on such results, additional highlights can be obtained regarding the best intervention logic which globally applies to a specific critical road scenario. Figure 7 evidences that the AEB intervention results in no substantial benefit for vehicle B in terms of IR , as reported in a case study by Vangi et al [42] where the [42], in case of "no intervention", "100% braking" (AEB) and "adaptive logic" for an ADAS implemented on board vehicle A (arrows represent the closing velocity vectors, starting from the point of impact) ego-vehicle (A) and the opponent (B) travel an intersection at 50 km/h; the scenario is similar to that of Fig. 6, the difference lying in the vehicles' position at the beginning of the criticality.…”

Section: Prospective Analyses-comparison Between Different Types Of A...supporting

confidence: 73%

“…Figure 5 summarizes input data and results in terms of CMI and V r deriving from activation (MS) or non-activation (RS) of an AEB implemented on board vehicle A. With the AEB, it is assumed that a maximum deceleration equal to 8 m/s 2 can be achieved by the ADAS device [42]. Vehicle A decelerates for approximately 0.4 s, reaching a longitudinal collision speed V i of 50 km/h and impacting Vehicle B in correspondence of the compartment.…”

Section: Prospective Analyses-comparison Between Different Types Of A...mentioning

confidence: 99%

“…6, the ADAS activation determines a change in the coordinates Fig. 4 Example model where IR is a function of V , impact zone and position of the occupant [42] toward areas characterized by low IR for vehicle A that sustains a frontal impact in both scenarios (from 27% to 7% according to Fig. 4); however, in the MS vehicle B undergoes a near side impact, so that IR associated with the MS is almost equal to that of the RS (27%).…”

Section: Prospective Analyses-comparison Between Different Types Of A...mentioning

confidence: 99%

See 2 more Smart Citations

Prospective and retrospective performance assessment of Advanced Driver Assistance Systems in imminent collision scenarios: the CMI-Vr approach

Gulino

Fiorentino²,

Vangi

2022

Eur. Transp. Res. Rev.

Self Cite

View full text Add to dashboard Cite

Structured abstract Introduction Prospective and retrospective performance assessment of Advanced Driver Assistance Systems (ADASs) is fundamental to pilot future enhancements for active safety devices. In critical road scenarios between two vehicles where ADAS activation enables collision mitigation only, currently available assessment methodologies rely on the reconstruction of the impact phase consequent to the specific intervention on braking and steering: the velocity change sustained by the vehicle in the collision ($$\Delta V$$ Δ V ) is retrieved, so that IR decrease for the vehicle occupants can be obtained by appropriate Injury Risk (IR) models. However, information regarding the ADAS performance is available only after the impact phase reconstruction and not just as when the criticality occurs in the pre-impact phase: the best braking and steering alternative cannot be immediately envisaged, since a direct correlation lacks between the braking/steering intervention and IR. Method This work highlights an ADAS performance assessment method based on the disaggregation of $$\Delta V$$ Δ V in the two pre-impact parameters closing velocity at collision ($$V_r$$ V r ) and impact eccentricity, represented by the Crash Momentum Index (CMI). Such a disaggregation leads to the determination of IR based solely on impact configuration between the vehicles, without directly considering the impact phase. The performance of diverse ADASs in terms of intervention logic are directly comparable based on the resulting impact configuration, associated with a single coordinate in the CMI-$$V_r$$ V r plane and a sole IR value as a consequence. Results The CMI-$$V_r$$ V r approach is employable for both purposes of prospective and retrospective performance assessment of ADAS devices. To illustrate the advantages of the methodology, a solution for prospective assessment based on the CMI-$$V_r$$ V r plane is initially proposed and applied to case studies: this provides direct suggestions regarding the most appropriate interventions on braking and steering for IR minimization, fundamental in the tuning or development phase of an ADAS. A method for retrospective assessment is ultimately contextualized in the EuroNCAP “Car-to-Car Rear moving” test for an Inter-Urban Autonomous Emergency Braking system, a device implemented on a significant portion of the circulating fleet. Conclusions Based on the evidenced highlights, it is demonstrated that the approach provides complementary information compared to well-established performance assessment methodologies in all stages of an ADAS life cycle, by suggesting a direct physical connection in the pre-impact phase between the possible ADAS interventions and the foreseeable injury outcomes.

show abstract

Section: Prospective Analyses-comparison Between Different Types Of A...supporting

confidence: 73%

Section: Prospective Analyses-comparison Between Different Types Of A...mentioning

confidence: 99%

Section: Prospective Analyses-comparison Between Different Types Of A...mentioning

confidence: 99%

See 1 more Smart Citation

Prospective and retrospective performance assessment of Advanced Driver Assistance Systems in imminent collision scenarios: the CMI-Vr approach

Gulino

Fiorentino²,

Vangi

2022

Eur. Transp. Res. Rev.

Self Cite

View full text Add to dashboard Cite

show abstract

“…With the evolution of advanced assisted driving system and automatic driving technology, the road accident probability is greatly reduced and the driving safety is improved [ 1 , 2 , 3 , 4 ]. As a key and challenging part of automatic driving and advanced assistant system, lane detection has also become a research hotspot [ 5 , 6 , 7 , 8 ].…”

Section: Introductionmentioning

confidence: 99%

Lane-GAN: A Robust Lane Detection Network for Driver Assistance System in High Speed and Complex Road Conditions

Liu

Wang

et al. 2022

Micromachines

View full text Add to dashboard Cite

Lane detection is an important and challenging part of autonomous driver assistance systems and other advanced assistance systems. The presence of road potholes and obstacles, complex road environments (illumination, occlusion, etc.) are ubiquitous, will cause the blur of images, which is captured by the vision perception system in the lane detection task. To improve the lane detection accuracy of blurred images, a network (Lane-GAN) for lane line detection is proposed in the paper, which is robust to blurred images. First, real and complex blur kernels are simulated to construct a blurred image dataset, and the improved GAN network is used to reinforce the lane features of the blurred image, and finally the feature information is further enriched with a recurrent feature transfer aggregator. Extensive experimental results demonstrate that the proposed network can get robust detection results in complex environments, especially for blurred lane lines. Compared with the SOTA detector, the proposed detector achieves a larger gain. The proposed method can enhance the lane detail features of the blurred image, improving the detection accuracy of the blurred lane effectively, in the driver assistance system in high speed and complex road conditions.

show abstract

“…Advanced driver assistance systems (ADAS) can play a key role [1][2][3] to reduce traffic accidents, which has a core module, namely lane detection, to keep the vehicles within the lane. 4 In recent years, a large number of lane detection methods have been proposed and applied in intelligent driving or ADAS.…”

Section: Introductionmentioning

confidence: 99%

A novel deep learning network for accurate lane detection in low-light environments

Song

Chen

Liu

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

Lane detection algorithms play a key role in Advanced Driver Assistance Systems (ADAS), which are however unable to achieve accurate lane recognition in low-light environments. This paper presents a novel deep network structure, namely LLSS-Net (low-light images semantic segmentation), to achieve accurate lane detection in low-light environments. The method integrates a convolutional neural network for low-light image enhancement and a semantic segmentation network for lane detection. The image quality is firstly improved by a low-light image enhancement network and lane features are then extracted using semantic segmentation. Fast lane clustering is finally performed by using the KD tree models. Cityscapes and Tusimple datasets are utilized to demonstrate the robustness of the proposed method. The experimental results show that the proposed method has an excellent performance for lane detection in low-light roads.

show abstract

Adaptive intervention logic for automated driving systems based on injury risk minimization

Cited by 7 publications

References 52 publications

Prospective and retrospective performance assessment of Advanced Driver Assistance Systems in imminent collision scenarios: the CMI-Vr approach

Prospective and retrospective performance assessment of Advanced Driver Assistance Systems in imminent collision scenarios: the CMI-Vr approach

Lane-GAN: A Robust Lane Detection Network for Driver Assistance System in High Speed and Complex Road Conditions

A novel deep learning network for accurate lane detection in low-light environments

Contact Info

Product

Resources

About