Automated Driving Vehicles

Yi, Kyongsu; Song, Bang‐Sup

doi:10.1007/978-3-030-75884-4_6

Cited by 1 publication

(4 citation statements)

References 79 publications

(88 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The strategies for collision avoidance are defined in Section II. The generated trajectories are based on lateral motion for lane change [24]. To avoid the high acceleration jerk, refer [24] suggests lane change motion with limited acceleration based on hyperbolic tangent path.…”

Section: ) Motion Predictionmentioning

confidence: 99%

“…The generated trajectories are based on lateral motion for lane change [24]. To avoid the high acceleration jerk, refer [24] suggests lane change motion with limited acceleration based on hyperbolic tangent path. However, this method is not considered roads with curvature.…”

Section: ) Motion Predictionmentioning

confidence: 99%

“…Choosing relevant scenarios from numerous cases is challenging. Furthermore, because the performance of automatic evasive steering systems can vary based on test scenarios, it's crucial to define a specific set of scenarios, termed a "scenario catalog," for the development and evaluation of AES [24].…”

Section: ) Simulation Of Safety-critical Scenariosmentioning

confidence: 99%

“…This method is also referred to as rasterized or vectorized High-Definition (HD) maps [22,23]. While the first and second approaches are in general sensitive to characteristics and performance of the environmental sensors, the SBEV approach may be more robust due to sensor fusion integrating the data coming from different sensors (e.g., radar, lidar, and camera) [22], [24]. With the advantage of the SBEV, a stacked set of current and past SBEV including lane mark and position of SVs are used as the network input to classify lane change intention [25].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Data-Driven Strategy Decision Integrating Convolution Neural Network With Threat Assessment and Motion Prediction for Automatic Evasive Steering

Lee,

Song

2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

In this paper, the strategy decision algorithm for automatic evasive steering (AES) integrating a convolution neural network (CNN) with a physics-based threat assessment is proposed. Five collision avoidance or mitigation strategies, including evasive steering, lane change, and steering to shoulder stop are considered for the strategy decision. Although there are many model-based or data-driven approaches for collision avoidance in the literature, a new decision method integrating data-driven classification based on CNN with both threat assessment and prediction techniques is proposed to improve reliability as well as accuracy. First, a set of abstracted images in a bird eye's view including the threat assessment and trajectory prediction information are generated. More specifically, a few collision indexes and interaction multiple model-unscented Kalman filter are used respectively for threat assessment and prediction. Once a stack of the images so called predicted semantic map corresponding to each collision avoidance strategy are generated, the decision classification based on CNN follows to choose an appropriate strategy for AES. Finally, the proposed decision algorithm is trained and validated through typical safe scenario data coming from field operation tests and safety-critical scenario data via simulations.

show abstract

Section: ) Motion Predictionmentioning

confidence: 99%

Section: ) Motion Predictionmentioning

confidence: 99%

Section: ) Simulation Of Safety-critical Scenariosmentioning

confidence: 99%