Autonomous cars controlled by an artificial intelligence are increasingly being integrated in the transport portfolio of cities, with strong repercussions for design and sustainability built environment. This paper sheds light on urban transition to autonomous transport, a threefold manner. First, we advance theoretical framework understand diffusion basis three interconnected factors: social attitudes, technological innovation politics. Second, draw upon in-depth survey conducted Dublin...

Connected and Autonomous Vehicles (CAVs) are expected to bring major transformations transport efficiency safety. Studies show a range of possible impacts, from worse CAVs at low penetration rates, significant improvements in both safety high rates loads. However, these studies tend explore separately, focus on one type road network, include only cars rather than other vehicle types. This paper presents comprehensive study impact safety, three types networks (urban, national, motorway),...

The growth of data generation capabilities, facilitated by advancements in communication and computation technologies, as well the rise Internet Things (IoT), results vast amounts that significantly enhance performance machine learning models. However, collecting all necessary to train accurate models is often unfeasible due privacy laws. Federated Learning (FL) evolved a collaborative approach for training without sharing private data. Unfortunately, several in-design vulnerabilities have...

Applications such as generator scheduling, household smart device transmission line overload management and microgrid islanding autonomy all play key roles in the grid ecosystem. Management of these applications could benefit from short-term load prediction, which has been successfully achieved on large-scale systems national grids. However, scale data for analysis is much smaller, similar to a single transformer, making prediction difficult. This paper examines several approaches day week...

In this letter, we study the energy efficiency (EE) optimization of unmanned aerial vehicles (UAVs) providing wireless coverage to static and mobile ground users. Recent multi-agent reinforcement learning approaches optimise system's EE using a 2D trajectory design, neglecting interference from nearby UAV cells. We aim maximize by jointly optimizing each UAV's 3D trajectory, number connected users, consumed, while accounting for interference. Thus, propose cooperative Multi-Agent...

Multi-agent reinforcement learning (MARL) is a widely researched technique for decentralised control in complex large-scale autonomous systems. Such systems often operate environments that are continuously evolving and where agents’ actions non-deterministic, so called inherently non-stationary environments. When there inconsistent results agents acting on such an environment, adapting challenging. In this article, we propose P-MARL, approach integrates prediction pattern change detection...

Improving the efficiency of smart grid, and in particular efficient integration energy from renewable sources, is key to sustainability electricity provision. In order optimize usage, demand response mechanisms are needed shift usage periods low demand, or high availability energy. this paper we propose a multi-agent approach that uses load forecasting for residential response. Electrical devices household controlled by reinforcement learning agents which, using information on current...

Shared mobility-on-demand systems can improve the efficiency of urban mobility through reduced vehicle ownership and parking demand. However, some issues in their implementations remain open, most notably issue rebalancing non-occupied vehicles to meet geographically uneven demand, as is, for example, case during rush hour. This is somewhat alleviated by prospect autonomous systems, where relocate themselves; however, proposed relocation strategies are still centralized assume all a part...

A Zero Energy Building (ZEB) has its net energy usage over a period of one year as zero, i.e., use is not larger than overall renewables generation. collection such ZEBs forms Community (ZEC). This paper addresses the problem sharing in community. different from previously addressed between buildings our focus on improvement community status, while traditionally research focused reducing losses due to transmission and storage, or achieving economic gains. We model this multi-agent...

Mobility-on-demand (MOD) systems consisting of shared autonomous vehicles (SAVs) are expected to improve the efficiency urban transportation through reduced vehicle ownership and parking demand. However, several issues related their implementation remain open, such as unifying ridesharing (RS) assignment with rebalancing (RB) unoccupied vehicles.

Large-scale, multi-agent systems are too complex for optimal control strategies to be known at design time and as a result good must learned runtime. Learning in such systems, particularly those with multiple objectives, takes considerable amount of because the size environment dependencies between goals. Transfer (TL) has been shown reduce learning single-agent, single-objective applications. It is process sharing knowledge two tasks called source target. The required have completed prior...

Due to steady urbanization, the electrical grid is facing significant changes in supply of resources as well type, scale, and patterns residential user demand. To ensure sustainability reliability electricity provision growing cities, a increase energy generated from renewable sources (e.g., wind, solar) required. However, much more variable intermittent than traditional supply, it depends on changing weather conditions. In order optimize usage, demand response (DR) techniques are being...

Various multi-agent decentralized approaches based on reinforcement learning (RL) have been proposed to increase scalability and real-time adaptiveness of urban traffic control (UTC) systems. In such approaches, light parameters are not pre-defined, but intelligent agents controlling the junctions learn suitable signal settings. order consider applications RL in commercial UTC products, they need enable fine-grained optimization multiple objectives be validated realistic simulations. This...

Variable Speed Limit (VSL) is a traffic control approach that optimises the mainstream on motorways. Reinforcement Learning to VSL has been shown achieve improvements in controlling bottleneck However, single-agent VSL, applied shorter motorway segment, can produce discontinuity flow by causing significant differences speeds between uncontrolled upstream and affected VSL. A multi-agent strategy be used overcome these problems assigning speed limits multiple sections enabling smoother...

Unmanned aerial vehicles (UAVs) are increasingly deployed to provide wireless connectivity static and mobile ground users in situations of increased network demand or points failure existing terrestrial cellular infrastructure. However, UAVs energy-constrained experience the challenge interference from nearby UAV cells sharing same frequency spectrum, thereby impacting system's energy efficiency (EE). Recent approaches focus on optimising EE by trajectory serving only neglecting users....

Large-scale agent-based systems are required to self-optimize towards multiple, potentially conflicting, policies of varying spatial and temporal scope. As a result, not all agents may be implementing at times, resulting in agent heterogeneity. share their operating environment, significant dependencies can arise between therefore policy implementations. To address self-optimization the presence heterogeneity, dependency lack global knowledge that is inherent large-scale decentralized...

Unmanned Aerial Vehicles (UAVs) are emerging as important users of next-generation cellular networks. By operating in the sky, UAV experience very different radio conditions than terrestrial users, due to factors such strong Line-of-Sight (LoS) channels (and interference) and Base Station (BS) antenna misalignment. As a consequence, UAVs may significant degradation their received quality service, particularly when they moving subject frequent handovers. The solution is allow be aware its...

The prevailing variable speed limit (VSL) systems as an effective strategy for traffic control on motorways have the disadvantage that they only work with static VSL zones. Under changing conditions, zones may perform suboptimally. Therefore, adaptive design of is required in scenarios where congestion characteristics vary widely over space and time. To address this problem, we propose a novel distributed spatial-temporal multi-agent (DWL-ST-VSL) approach capable dynamically adjusting length...

This article describes Distributed W-Learning (DWL), a reinforcement learning-based algorithm for collaborative agent-based optimization of pervasive systems. DWL supports towards multiple heterogeneous policies and addresses the challenges arising from heterogeneity agents that are charged with implementing them. learns exploits dependencies between to improve overall system performance. Instead always executing locally-best action, learn how their actions affect immediate neighbors execute...