Advanced Adaptive Cruise Control Based on Collision Risk Assessment

Woo, Hanwool; Ji, Yonghoon; Tamura, Yusuke; Kuroda, Yasuhide; Sugano, Takashi; Yamamoto, Yasunori; Yamashita, Atsushi; Asama, Hajime

doi:10.1109/itsc.2018.8569759

Cited by 8 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…He et al [25] used a Dynamic Bayesian Network (DBN) for predicting lane change and lane-keeping intention on highways, and a cosine function for trajectory estimation. Similarly, Woo et al [26] predicted maneuver intentions using multi-class Support Vector Machine (SVM), and applied a sinusoidal and potential field functions to generate, respectively, lateral and longitudinal trajectories. Sinusoidal functions are often used for lane change path estimation [26]- [31].…”

Section: A Analytical Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A Review on Intention-aware and Interaction-aware Trajectory Prediction for Autonomous Vehicles

Gomes¹,

Wolf²

2022

Preprint

View full text Add to dashboard Cite

This paper presents a literature review on Intention-aware and Interaction-aware Trajectory Prediction for Autonomous Vehicle, which covers primary studies since 2008. The research questions try to understand how intention/interaction improves prediction, and what are the techniques and technologies employed in trajectory prediction.<div>Moreover, this paper is under review in the IEEE Transactions on Intelligent Transportation Systems.</div>

show abstract

Section: A Analytical Methodsmentioning

confidence: 99%

“…Similarly, Woo et al [26] predicted maneuver intentions using multi-class Support Vector Machine (SVM), and applied a sinusoidal and potential field functions to generate, respectively, lateral and longitudinal trajectories. Sinusoidal functions are often used for lane change path estimation [26]- [31].…”

Section: A Analytical Methodsmentioning

confidence: 99%

A Review on Intention-aware and Interaction-aware Trajectory Prediction for Autonomous Vehicles

Gomes¹,

Wolf²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Higgs et al reported the limitation of the Wiedemann car-following model, which is one of the microscopic models, caused by the individual differences of drivers [7]. Previously suggested methods, including the ones suggested in our previous work, do not consider the individual difference in characteristics [9], leading to the loss of the trajectory prediction accuracy. In addition, the methods with the predetermined values proposed in the previous studies cannot handle the changes in driving conditions.…”

Section: Introductionmentioning

confidence: 95%

Advanced Adaptive Cruise Control Based on Operation Characteristic Estimation and Trajectory Prediction

et al. 2019

Self Cite

View full text Add to dashboard Cite

In this paper, we propose an advanced adaptive cruise control to evaluate the collision risk between adjacent vehicles and adjust the distance between them seeking to improve driving safety. As a solution for preventing crashes, an autopilot vehicle has been considered. In the near future, the technique to forecast dangerous situations and automatically adjust the speed to prevent a collision can be implemented to a real vehicle. We have attempted to realize the technique to predict the future positions of adjacent vehicles. Several previous studies have investigated similar approaches; however, these studies ignored the individual characteristics of drivers and changes in driving conditions, even though the prediction performance largely depends on these characteristics. The proposed method allows estimating the operation characteristics of each driver and applying the estimated results to obtain the trajectory prediction. Then, the collision risk is evaluated based on such prediction. A novel advanced adaptive cruise control, proposed in this paper, adjusts its speed and distance from adjacent vehicles accordingly to minimize the collision risk in advance. In evaluation using real traffic data, the proposed method detected lane changes with 99.2% and achieved trajectory prediction error of 0.065 m, on average. In addition, it was demonstrated that almost 35% of the collision risk can be decreased by applying the proposed method compared to that of human drivers.

show abstract

“…Lane change is a typical driving maneuver that can be performed at a varied level of risk, and a significant percentage of all crashes happen due to the erroneous execution of it [9]. Risk in lane changes was studied mostly from the perspective of objective collision risk minimization [11], [15]. A lane change dataset with manually annotated subjective risk labels made it possible to approach this problem from the perspective of supervised learning.…”

Section: A Risk Studiesmentioning

confidence: 99%

Risky Action Recognition in Lane Change Video Clips using Deep Spatiotemporal Networks with Segmentation Mask Transfer

Yurtsever

Liu

Lambert

et al. 2019

2019 IEEE Intelligent Transportation Systems Conference (ITSC)

View full text Add to dashboard Cite

Advanced driver assistance and automated driving systems rely on risk estimation modules to predict and avoid dangerous situations. Current methods use expensive sensor setups and complex processing pipeline, limiting their availability and robustness. To address these issues, we introduce a novel deep learning based action recognition framework for classifying dangerous lane change behavior in short video clips captured by a monocular camera. We designed a deep spatiotemporal classification network that uses pre-trained state-of-the-art instance segmentation network Mask R-CNN as its spatial feature extractor for this task. The Long-Short Term Memory (LSTM) and shallower final classification layers of the proposed method were trained on a semi-naturalistic lane change dataset with annotated risk labels. A comprehensive comparison of state-of-the-art feature extractors was carried out to find the best network layout and training strategy. The best result, with a 0.937 AUC score, was obtained with the proposed network. Our code and trained models are available open-source 1 .

show abstract

Advanced Adaptive Cruise Control Based on Collision Risk Assessment

Cited by 8 publications

References 10 publications

A Review on Intention-aware and Interaction-aware Trajectory Prediction for Autonomous Vehicles

A Review on Intention-aware and Interaction-aware Trajectory Prediction for Autonomous Vehicles

Advanced Adaptive Cruise Control Based on Operation Characteristic Estimation and Trajectory Prediction

Risky Action Recognition in Lane Change Video Clips using Deep Spatiotemporal Networks with Segmentation Mask Transfer

Contact Info

Product

Resources

About