No abstract

Future Intelligent Transportation Systems (ITS) can improve on-road safety and transportation efficiency and vehicular networks (VNs) are essential to enable ITS applications via information sharing. The development of 5G introduces new technologies providing improved support for connected vehicles through highly dynamic heterogeneous networks. Machine Learning (ML) can capture the high dynamics of VNs but the distributed data cause new challenges for ML and requires distributed solutions. Federated learning (FL), a distributed ML framework, gives a distributed ML framework while ensuring information privacy protection and is an exciting area to explore in VNs. This article provides a detailed summary of recent FL applications in VNs and gives insights on current research challenges. The included research topics are resource management, performance optimization and applications based on VNs.

Currently for balance recovery, humans outperform humanoid robots that used hand-designed controllers. This study aims to close this gap by finding control principles which are shared across all recovery strategies. We do this by formulating experiments to test human strategies and quantify criteria for identifying strategies. A minimum jerk control principle is shown to accurately recreate human CoM recovery trajectories. Using this principle, we formulate a Model-Predictive Control (MPC) for the use in floating base systems (eg legged robots). The feasibility of generated motions from the MPC for implementation on the real robot is then validated using an Inverted Pendulum Model. Finally, we demonstrate improved capability over humans by tuning the parameters for time-optimal recovery performance.

An Intelligent Transportation System (ITS) application requires vehicles to be connected to each other and to roadside units to share information, thus reducing fatalities and improving traffic congestion. Vehicular Ad hoc Networks (VANETs) is one of the main forms of network designed for ITS in which information is broadcasted amongst vehicular nodes. However, the broadcast reliability in VANETs face a number of challenges -dynamic routing being one of the major issues. Clustering, a technique used to group nodes based on certain criteria, has been suggested as a solution to this problem. This paper gives a summary of the core criteria of some of the clustering algorithms issues along with a performance comparison and a development evolution roadmap, in an attempt to understand and differentiate different aspects of the current research and suggest future research insights.

For vehicle-to-network communications, handover (HO) management enables vehicles to maintain the connection with the network while transiting through coverage areas of different base stations (BSs). However, the high mobility of vehicles means shorter connection periods with each BS that leads to frequent HOs, hence raises the necessity for optimal HO decision making for high quality infotainment services. Machine learning is capable of capturing underlying pattern via data driven methods to find optimal solutions to complex problems, and much learning-based HO optimization research has been conducted focusing on specific network setups. However, attention still needs to be paid to the actual deployment aspect and standardized datasets or simulation environments for evaluation. This paper proposes a deep reinforcement learning-based HO algorithm using the input parameters that are configurable in the existing measurement report of cellular networks. The performance of the proposed algorithm is evaluated using the well-known network simulator ns-3 with its official LTE module. A realistic network setup in the city center of Glasgow (UK) is configured with vehicle trajectories generated by the routes mobility model using Google Maps Directions API. Evaluation results reveal that the proposed algorithm significantly outperforms the A3 RSRP baseline with an average of 25.72% packet loss reduction per HO, suggesting significant improvement in quality of service of phone call and video streaming, etc. The proposed algorithm also has a small implementation cost compared to some state-of-the-art and should be deployed by a software update to a local BS controller.