Cooperative Perception with Deep Reinforcement Learning for Connected Vehicles

Aoki, Sadao; Higuchi, Tetsuya; Altintas, Onur

doi:10.1109/iv47402.2020.9304570

Cited by 82 publications

(44 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With massive data storage and data sharing in the digital space, multiple vehicle-related microservices can be enabled, such as the ones requiring cooperation among multiple connected vehicles: cooperative localization [45], cooperative perception [46], cooperative planning [47], and cooperative control [14]. Additionally, microservices that need time-series data can also be benefited from this MDT framework, where one typical example is predictive maintenance: Based on modeling and simulation of the time-series vehicle data that is sampled from the Vehicle block in the physical space and stored in the Vehicle Digital Twin, the learning process can be conducted in the digital space and predictions can be made regarding potential failures of vehicle components at a future time [48].…”

Section: Digital Spacementioning

confidence: 99%

Mobility Digital Twin: Concept, Architecture, Case Study, and Future Challenges

Wang¹

2022

Preprint

View full text Add to dashboard Cite

<div>A Digital Twin is a digital replica of a living or non-living physical entity, and this emerging technology has attracted extensive attention from different industries during the past decade. Although a few Digital Twin studies have been conducted in the transportation domain very recently, there is no systematic research with a holistic framework connecting various mobility entities together. In this study, by leveraging both connected vehicle technology and cloud computing, an Mobility Digital Twin (MDT) framework is developed, which consists of three building blocks in the physical space (namely Human, Vehicle, and Traffic), and their associated Digital Twins in the digital space. The cloud architecture is built with Amazon Web Services (AWS) to accommodate the proposed MDT framework and to implement its digital functionalities of storage, modeling, learning, simulation, and prediction. The effectiveness of the MDT framework is shown through the case studies of three digital building blocks with their key microservices: the Human Digital Twin with user management and driver type classification, the Vehicle Digital Twin with cloud-based Advanced Driver-Assistance Systems (ADAS), and the Traffic Digital Twin with traffic flow monitoring and variable speed limit.</div>

show abstract

Section: Digital Spacementioning

confidence: 99%

Mobility Digital Twin: Concept, Architecture, Case Study, and Future Challenges

Wang¹

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Cooperative perception is foreseen as a potential approach that overcomes and improves the sensing-based technology limitations of CVs by allowing vehicles to exchange their local raw or processed sensor information via V2V communication to improve their visibility and field of vision. In a dense environment, vehicles may redundantly transmit the same perceived data to nearby vehicles, which results in a reduction in V2V communication reliability, inefficient use of network resources, and high network loads that lead to emergency data or critical information being lost or delayed and hence jeopardizing road safety [149,150]. Developing intelligent schemes to mitigate redundant data and improve network loads and scalability while utilizing network resources efficiently are promising research directions.…”

Section: Opportunitiesmentioning

confidence: 99%

Connected Vehicles: Technology Review, State of the Art, Challenges and Opportunities

Abdelkader

Elgazzar

Khamis

2021

Sensors

View full text Add to dashboard Cite

In an effort to reach accident-free milestones or drastically reduce/eliminate road fatalities rates and traffic congestion and to create disruptive, transformational mobility systems and services, different parties (e.g., automakers, universities, governments, and road traffic regulators) have collaborated to research, develop, and test connected vehicle (CV) technologies. CVs create new data-rich environments and are considered key enablers for many applications and services that will make our roads safer, less congested, and more eco-friendly. A deeper understanding of the CV technologies will pave the way to avoid setbacks and will help in developing more innovative applications and breakthroughs. In the CV paradigm, vehicles become smarter by communicating with nearby vehicles, connected infrastructure, and the surroundings. This connectivity will be substantial to support different features and systems, such as adaptive routing, real-time navigation, and slow and near real-time infrastructure. Further examples include environmental sensing, advanced driver-assistance systems, automated driving systems, mobility on demand, and mobility as a service. This article provides a comprehensive review on CV technologies including fundamental challenges, state-of-the-art enabling technologies, innovative applications, and potential opportunities that can benefit automakers, customers, and businesses. The current standardization efforts of the forefront enabling technologies, such as Wi-Fi 6 and 5G-cellular technologies are also reviewed. Different challenges in terms of cooperative computation, privacy/security, and over-the-air updates are discussed. Safety and non-safety applications are described and possible future opportunities that CV technology brings to our life are also highlighted.

show abstract

“…The challenge is to obtain an accurate estimation of the value of the information. This challenge has been partially addressed in a recent study by the same authors in [18] where they propose the use of deep reinforcement learning to select the data to transmit. A similar concept was proposed in [19] where authors present a method for each vehicle to dynamically adapt the message transmission rate taking into account the area covered with their sensors and that is not covered by nearby vehicles.…”

Section: State Of the Artmentioning

confidence: 99%

Cooperative Perception for Connected and Automated Vehicles: Evaluation and Impact of Congestion Control

2020

View full text Add to dashboard Cite

Automated vehicles make use of multiple sensors to detect their surroundings. Sensors have significantly improved over the years but still face challenges due to the presence of obstacles or adverse weather conditions, among others. Cooperative or collective perception has been proposed to help mitigate these challenges through the exchange of sensor data among vehicles using V2X (Vehicle-to-Everything) communications. Recent studies have shown that cooperative perception can complement on-board sensors and increase the vehicle's awareness beyond its sensors field of view. However, cooperative perception significantly increases the amount of information exchanged by vehicles which can degrade the V2X communication performance and ultimately the effectiveness of cooperative perception. In this context, this study conducts first a dimensioning analysis to evaluate the impact of the sensors' characteristics and the market penetration rate on the operation and performance of cooperative perception. The study then investigates the impact of congestion control on cooperative perception using the Decentralized Congestion Control (DCC) framework defined by ETSI. The study demonstrates that congestion control can negatively impact the perception and latency of cooperative perception if not adequately configured. In this context, this study demonstrates for the first time that the combination of congestion control functions at the Access and Facilities layers can improve the perception achieved with cooperative perception and ensure a timely transmission of the information. The results obtained demonstrate the importance of an adequate configuration of DCC for the development of connected and automated vehicles.

show abstract

Cooperative Perception with Deep Reinforcement Learning for Connected Vehicles

Cited by 82 publications

References 17 publications

Mobility Digital Twin: Concept, Architecture, Case Study, and Future Challenges

Mobility Digital Twin: Concept, Architecture, Case Study, and Future Challenges

Connected Vehicles: Technology Review, State of the Art, Challenges and Opportunities

Cooperative Perception for Connected and Automated Vehicles: Evaluation and Impact of Congestion Control

Contact Info

Product

Resources

About