Driverless Car: Autonomous Driving Using Deep Reinforcement Learning in Urban Environment

Fayjie, Abdur Razzaq; Hossain, Sabir; Doukhi, Oualid; Lee, Deok Jin

doi:10.1109/urai.2018.8441797

Cited by 110 publications

(52 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…If there is a traffic jam or accident on the route, the driver will be notified. By combining the transport infrastructure, IoT can create a collaborative system that can show how people reach their destinations safely and on time [22]. Figure 2 shows the traffic light system connected to the traffic camera, the gateway, and the wireless smart mesh IP sensor network of the vehicles to form an IoT-based traffic light management complex.…”

Section: )mentioning

confidence: 99%

Core Orientations for 4.0 Technology Application on the Development Strategy of Intelligent Transportation System in Vietnam

Nguyen¹

2020

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

The Industrial Revolution 4.0 is having a substantial impact on countries in many fields such as industry structure, human resource needs, production management systems. Transportation is a high-community industry, so it will also be strongly affected by this industrial revolution. The achievements of Industry 4.0 can reshape the nation's transportation systems. That poses challenges for transportation managers to change the operating model as well as the transport-related policies. Building and developing intelligent transport systems (ITS) is attracting considerable attention, research, and investment to serve the development of transportation systems in the world as well as in Vietnam. Vietnam is gradually building Intelligent Governments, Intelligent Cities, and the Intelligent Transportation system will be an essential part of that development and entering the stage of access to the 4.0 industrial revolution, building a smart city applying 4.0 technology to bring convenience, friendliness, and safety to people in each city. As a step ahead, the Transport industry gradually realizes its dream of developing intelligent transportation in developed cities like Hanoi and Ho Chi Minh City. The paper presents an overview of the impact of Industry 4.0 on key aspects of the transport sector in general and ITS development strategy in Vietnam. On that basis, state-of-the-art technologies-technology 4.0 are proposed to meet the demands of intelligent transportation development. The paper will show that 4.0 technology is the key to the development of ITS in Vietnam.

show abstract

Section: )mentioning

confidence: 99%

Core Orientations for 4.0 Technology Application on the Development Strategy of Intelligent Transportation System in Vietnam

Nguyen¹

2020

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

show abstract

“…This is different from the high-resolution perception result, but the idea of using grid space is still important. Some researchers have used raw sensor input and deep reinforcement learning [96,97] with images or point clouds as input [97]. Liu et al [98] further combined deep reinforcement learning with supervised deep learning to make driving decisions.…”

Section: The Driving Space In the Reinforcement Learning Decisionmentioning

confidence: 99%

Driving Space for Autonomous Vehicles

et al. 2019

View full text Add to dashboard Cite

Driving space for autonomous vehicles (AVs) is a simplified representation of real driving environments that helps facilitate driving decision processes. Existing literatures present numerous methods for constructing driving spaces, which is a fundamental step in AV development. This study reviews the existing researches to gain a more systematic understanding of driving space and focuses on two questions: how to reconstruct the driving environment, and how to make driving decisions within the constructed driving space. Furthermore, the advantages and disadvantages of different types of driving space are analyzed. The study provides further understanding of the relationship between perception and decision-making and gives insight into direction of future research on driving space of AVs.

show abstract

“…The CNN detects, recognizes and infers information from the sensors to make a decision with the data being depth information rather than RGB data as in Reference [19]. To speed up the process of training the network, the paper proposes using simulations to create training data rather than real driving data which may take many hours to collect, a strategy also used in many other papers, such as [21,22].…”

Section: Deep Learning Architectures For Autonomous Drivingmentioning

confidence: 99%

“…Advances of deep reinforcement learning with deep Q learning networks, has been proposed for driverless vehicle control [21][22][23]. A network consisting of three convolutional layers and four dense layers was created in Reference [21], and tested in a simulated urban environment made by Unity Game Engine [24]. It was used with a deep Q learning network, meaning a reward system was put in place with the aim to teach the agent to move forward, when safe, and not to hit other objects.…”

Section: Deep Learning Architectures For Autonomous Drivingmentioning

confidence: 99%

“…A study by Isele et al [23] firstly looked at how deep reinforcement learning methods compare to the Time To Collision (TTC) method in learning how to navigate unsignalized intersections using four metrics for analysis: Success percentage, percentage of collisions, the average time and average breaking time. Two deep Q networks (DQNs), such as those from References [21] and [22] were created. One was based on 'sequential actions' and the other on 'time-to-go'.…”

Section: Situational Intelligence For Driverless Vehicles In Intersecmentioning

confidence: 99%

See 1 more Smart Citation

Learning Control Policies of Driverless Vehicles from UAV Video Streams in Complex Urban Environments

2019

View full text Add to dashboard Cite

The way we drive, and the transport of today are going through radical changes. Intelligent mobility envisions to improve the efficiency of traditional transportation through advanced digital technologies, such as robotics, artificial intelligence and Internet of Things. Central to the development of intelligent mobility technology is the emergence of connected autonomous vehicles (CAVs) where vehicles are capable of navigating environments autonomously. For this to be achieved, autonomous vehicles must be safe, trusted by passengers, and other drivers. However, it is practically impossible to train autonomous vehicles with all the possible traffic conditions that they may encounter. The work in this paper presents an alternative solution of using infrastructure to aid CAVs to learn driving policies, specifically for complex junctions, which require local experience and knowledge to handle. The proposal is to learn safe driving policies through data-driven imitation learning of human-driven vehicles at a junction utilizing data captured from surveillance devices about vehicle movements at the junction. The proposed framework is demonstrated by processing video datasets captured from uncrewed aerial vehicles (UAVs) from three intersections around Europe which contain vehicle trajectories. An imitation learning algorithm based on long short-term memory (LSTM) neural network is proposed to learn and predict safe trajectories of vehicles. The proposed framework can be used for many purposes in intelligent mobility, such as augmenting the intelligent control algorithms in driverless vehicles, benchmarking driver behavior for insurance purposes, and for providing insights to city planning.

show abstract

Driverless Car: Autonomous Driving Using Deep Reinforcement Learning in Urban Environment

Cited by 110 publications

References 6 publications

Core Orientations for 4.0 Technology Application on the Development Strategy of Intelligent Transportation System in Vietnam

Core Orientations for 4.0 Technology Application on the Development Strategy of Intelligent Transportation System in Vietnam

Driving Space for Autonomous Vehicles

Learning Control Policies of Driverless Vehicles from UAV Video Streams in Complex Urban Environments

Contact Info

Product

Resources

About