Multimodel Approach to Personalized Autonomous Adaptive Cruise Control

Bolduc, Andrew Phillip; Guo, Longxiang; Jia, Yunyi

doi:10.1109/tiv.2019.2904419

Cited by 32 publications

(16 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conversely, many researchers have assembled their own vehicle controllers from in situ measurements of human drivers made possible by the miniaturization of dGPS and sensor suites (7,8). Lastly, with V2X communication now available and its operational properties becoming well known, detailed examination can be undertaken of aspects of cooperative systems, such as operational policies (9)(10)(11), latency and the effect of message corruption on overall efficiency (12) and interaction with traffic signals (13,14).…”

mentioning

confidence: 99%

Microscopic Modeling of the Effects of Autonomous Vehicles on Motorway Performance

Mesionis

Brackstone

Gravett

2020

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

Autonomous vehicles (AVs) have been the subject of extensive research in recent years and are expected to completely transform the operation of transport networks and revolutionize the automotive industry in the coming decades. Modeling detailed interactions among vehicles with varying levels of penetration rates is essential for evaluating the potential effects. One such investigation is being performed within the ‘HumanDrive’ Project in the U.K. This work has required the development of a behavioral model that incorporates microscopic level interactions and has been based on a pre-existing adaptive cruise control and lane-changing model that has been adapted to better replicate the limitations of AVs and allow the investigation of differing levels of intelligence or assertiveness. The model has been implemented on the M1 Motorway near Sheffield in the U.K. This has allowed the investigation of the effects of AVs on the operation of a real network under various traffic conditions where the overall effects may be revealed, both as advantages to AV drivers, and potentially disadvantages to non-AV traffic. Additionally, it has been possible to examine how these affect junction operations and net emissions. Preliminary results have allowed us to quantify the positive effects of AVs which increase with the penetration. However, it is clear that there are points of inflection where benefits start to slow. It is at these (high) penetration rates that initial operational assumptions may become increasingly stretched and additional infrastructure and cooperative systems are likely to have to become prevalent.

show abstract

mentioning

confidence: 99%

Microscopic Modeling of the Effects of Autonomous Vehicles on Motorway Performance

Mesionis

Brackstone

Gravett

2020

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

show abstract

“…Thus, AD models should be made mature enough to be universally trusted and adoptable at large scales. Further, personalization (such as preliminary explorations for cruise control [157] and lane departure [158], [159]) can be another interesting research direction for users to adjust their preferences in terms of safety, speed limit, available features, and cost.…”

Section: A) Energy-friendly Convolutional Neural Network (Cnns)mentioning

confidence: 99%

Deep Learning for Safe Autonomous Driving: Current Challenges and Future Directions

Muhammad

Ullah

Lloret

et al. 2021

IEEE Trans. Intell. Transport. Syst.

325

View full text Add to dashboard Cite

Advances in information and signal processing technologies have a significant impact on autonomous driving (AD), improving driving safety while minimizing the efforts of human drivers with the help of advanced artificial intelligence (AI) techniques. Recently, deep learning (DL) approaches have solved several real-world problems of complex nature. However, their strengths in terms of control processes for AD have not been deeply investigated and highlighted yet. This survey highlights the power of DL architectures in terms of reliability and efficient real-time performance and overviews state-of-theart strategies for safe AD, with their major achievements and limitations. Furthermore, it covers major embodiments of DL along the AD pipeline including measurement, analysis, and execution, with a focus on road, lane, vehicle, pedestrian, drowsiness detection, collision avoidance, and traffic sign detection through sensing and vision-based DL methods. In addition, we discuss on the performance of several reviewed methods by using different evaluation metrics, with critics on their pros and cons. Finally, this survey highlights the current issues of safe DL-based AD with a prospect of recommendations for future research, rounding up a reference material for newcomers and researchers willing to join this vibrant area of Intelligent Transportation Systems.

show abstract

“…The feature weight vector can be updated by the normalized gradient descent (NGD) method [11] as shown in (8)…”

Section: B Inverse Reinforcement Learning (Irl)mentioning

confidence: 99%

“…Existing studies use different mathematical models that mimic a human driver's driving behaviors to provide personalized ACC (PACC) in the dynamic traffic scenarios [6][7][8][9]. In [8], the authors propose a multimodel PACC approach to learn drivers' preferred driving behaviors such as highway driving preferences by extracting the numerical driving style indicators from driving data and to execute driver-specific control actions by using the model predictive M. F. Ozkan and Y. Ma are with the Department of Mechanical Engineering, Texas Tech University (e-mail: mehmet.ozkan@ttu.edu and yao.ma@ttu.edu). control (MPC) design.…”

Section: Introductionmentioning

confidence: 99%

Personalized Adaptive Cruise Control and Impacts on Mixed Traffic

Ozkan

2021

2021 American Control Conference (ACC)

View full text Add to dashboard Cite

This paper presents a personalized adaptive cruise control (PACC) design that can learn human driver behavior and adaptively control the semi-autonomous vehicle (SAV) in the car-following scenario, and investigates its impacts on mixed traffic. In mixed traffic where the SAV and humandriven vehicles share the road, the SAV's driver can choose a PACC tuning that better fits the driver's preferred driving strategies. The individual driver's preferences are learned through the inverse reinforcement learning (IRL) approach by recovering a unique cost function from the driver's demonstrated driving data that best explains the observed driving style. The proposed PACC design plans the motion of the SAV by minimizing the learned unique cost function considering the short preview information of the preceding human-driven vehicle. The results reveal that the learned driver model can identify and replicate the personalized driving behaviors accurately and consistently when following the preceding vehicle in a variety of traffic conditions. Furthermore, we investigated the impacts of the PACC with different drivers on mixed traffic by considering time headway, gap distance, and fuel economy assessments. A statistical investigation shows that the impacts of the PACC on mixed traffic vary among tested drivers due to their intrinsic driving preferences.

show abstract

Multimodel Approach to Personalized Autonomous Adaptive Cruise Control

Cited by 32 publications

References 28 publications

Microscopic Modeling of the Effects of Autonomous Vehicles on Motorway Performance

Microscopic Modeling of the Effects of Autonomous Vehicles on Motorway Performance

Deep Learning for Safe Autonomous Driving: Current Challenges and Future Directions

Personalized Adaptive Cruise Control and Impacts on Mixed Traffic

Contact Info

Product

Resources

About