Assisted driving system based on federated reinforcement learning

Article Computer Science, Hardware & Architecture

What are the actual needs of?

Menghan Hu et al.

Summary: In this editorial, the guest editor invited two visually impaired individuals to discuss the real needs of the visually impaired.

DISPLAYS (2023)

Add to Collection

Article Engineering, Civil

Survey of Deep Reinforcement Learning for Motion Planning of Autonomous Vehicles

Szilard Aradi

Summary: Academic research in the field of autonomous vehicles has gained popularity in recent years, covering various topics such as sensor technologies, communication, safety, decision making, and control. Artificial Intelligence and Machine Learning methods have become integral parts of this research. Motion planning, with a focus on strategic decision-making, trajectory planning, and control, has also been studied. This article specifically explores Deep Reinforcement Learning (DRL) as a field within Machine Learning. The paper provides insights into hierarchical motion planning and the basics of DRL, including environment modeling, state representation, perception models, reward mechanisms, and neural network implementation. It also discusses vehicle models, simulation possibilities, and computational requirements. The paper surveys state-of-the-art solutions, categorized by different tasks and levels of autonomous driving, such as car-following, lane-keeping, trajectory following, merging, and driving in dense traffic. Lastly, it raises open questions and future challenges.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2022)

Add to Collection

Article Computer Science, Cybernetics

FedVCP: A Federated-Learning-Based Cooperative Positioning Scheme for Social Internet of Vehicles

Xiangjie Kong et al.

Summary: This article proposes a vehicle cooperative positioning system based on federated learning, utilizing the potential of social Internet of Things and collaborative edge computing to provide high-precision positioning correction while ensuring user privacy.

IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS (2022)

Add to Collection

Article Engineering, Electrical & Electronic

Federated Learning With Non-IID Data in Wireless Networks

Zhongyuan Zhao et al.

Summary: This paper studies federated learning with non-IID data in wireless networks and proposes a federated averaging scheme and a joint optimization algorithm to reduce distribution divergence, maintaining a balance between model accuracy and cost. Simulation results show significant performance gains with a small price of latency and energy consumption.

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS (2022)

Add to Collection

Article Computer Science, Information Systems

WSCC: A Weight-Similarity-Based Client Clustering Approach for Non-IID Federated Learning

Pu Tian et al.

Summary: In this article, a weight-similarity-based client clustering approach is proposed to address the issue of non-IID data in IoT systems. The approach uses an affinity-propagation-based method with the cosine distance of the client's weight parameters to dynamically cluster clients, resulting in improved accuracy compared to existing federated learning schemes.

IEEE INTERNET OF THINGS JOURNAL (2022)

Add to Collection

Article Computer Science, Hardware & Architecture

Usage patterns of Head-mounted vision assistance app as compared to handheld video Magnifier

Gang Luo et al.

Summary: This study aims to investigate whether head-mounted vision assistance devices can address different visual needs. By collecting usage data from tens of thousands of global users over 8 months, the results show that head-mounted Supervision Goggles are used less frequently compared to handheld Supervision Magnifier, and the viewing targets differ between the two types of devices.

DISPLAYS (2022)

Add to Collection

Article Automation & Control Systems

A Federated Learning-Based License Plate Recognition Scheme for 5G-Enabled Internet of Vehicles

Xiangjie Kong et al.

Summary: This article proposes a federated learning-based license plate recognition framework (FedLPR) to solve the issues of centralized training of models and large number of parameters for deployment on mobile devices. By harnessing data on individuals' mobile devices and training models based on federated learning, FedLPR achieves high accuracy and acceptable communication cost while preserving user privacy.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Industrial Federated Topic Modeling

Di Jiang et al.

Summary: Probabilistic topic modeling is widely used in industrial applications, but the challenge lies in training high-quality models with sensitive or proprietary data. Industrial Federated Topic Modeling (iFTM) proposes a collaborative framework to address this challenge, balancing data scarcity and privacy concerns. Experimental results demonstrate the superiority of iFTM over traditional topic modeling approaches.

ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY (2021)

Add to Collection

Article Computer Science, Information Systems

A mutual information based federated learning framework for edge computing networks

Naiyue Chen et al.

Summary: With the advancement of artificial intelligence technology, federated learning has become a key technology to protect user privacy and solve the data island problem, but there is a challenge of malicious node attacks.

COMPUTER COMMUNICATIONS (2021)

Add to Collection

Article Engineering, Civil

Joint Computing and Caching in 5G-Envisioned Internet of Vehicles: A Deep Reinforcement Learning-Based Traffic Control System

Zhaolong Ning et al.

Summary: This paper develops an intent-based traffic control system for Connected Vehicles in the 5G era, utilizing Deep Reinforcement Learning to dynamically orchestrate edge computing and content caching to improve the profits of Mobile Network Operators. Experimental results based on real traffic data demonstrate the efficiency and performance of the designed system.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2021)

Add to Collection

Article Engineering, Civil

Towards Federated Learning in UAV-Enabled Internet of Vehicles: A Multi-Dimensional Contract-Matching Approach

Wei Yang Bryan Lim et al.

Summary: By utilizing Federated Learning, collaborative Machine Learning among independent DaaS providers can be achieved while preserving privacy for the development of IoV applications. The study leverages the self-revealing properties of multi-dimensional contracts to ensure truthful reporting of UAV types, while also considering multiple sources of heterogeneity.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2021)

Add to Collection

Article Ophthalmology

Home-Use Evaluation of a Wearable Collision Warning Device for Individuals With Severe Vision Impairments A Randomized Clinical Trial

Shrinivas Pundlik et al.

Summary: This study evaluated the effect of a collision warning device on the number of contacts experienced by blind and visually impaired individuals. The results showed that the device significantly reduced the rate of contacts with obstacles in daily mobility, suggesting its potential to enhance habitual mobility aids for visually impaired individuals.

JAMA OPHTHALMOLOGY (2021)

Add to Collection

Article Computer Science, Information Systems

Reward Shaping Based Federated Reinforcement Learning

Yiqiu Hu et al.

Summary: FRS is a general federated reinforcement learning framework that utilizes reward shaping as shared information among different clients to improve training speed and policy quality while protecting task privacy.

IEEE ACCESS (2021)

Add to Collection

Article Engineering, Electrical & Electronic