School of Computer Science and Statistics
In reinforcement learning, agents usually process all inputs equally, even if some are more relevant than others. This project will explore the use of a simplified attention mechanism in a car-following environment, allowing a vehicle to focus more on important neighbours (such as the vehicle directly in front). Students will compare the performance of Multi-Agent … Read more
Reinforcement learning agents often behave conservatively or inconsistently at the start of training, leading to the so-called cold-start problem. This project will simulate a simple traffic scenario and train a multi-agent RL-based autonomous vehicle controller. Students will then explore strategies to reduce cold-start effects, such as imitation learning for initialization or reusing past experiences, and … Read more
Before fully autonomous driving is achieved, connected autonomous vehicles (CAVs) will operate for a certain period in mixed traffic, which includes both CAVs and human-driven vehicles (HDVs). The dynamic and interactive conditions in mixed traffic scenarios renders CAV decision making particularly challenging. This project will investigate the use of Graph Convolutional Deep Reinforcement Learning for … Read more
This project will explore applications of Reinforcement Learning (RL) for a real-world application in the railway domain. To address increasing demand, railway providers aim to increase traffic density on the existing network. However, dense traffic can result in delayed trains and infrastructure disruptions, which can impact planned trips in a large part of the network … Read more
Poor travel-time reliability, meaning that travel times for the same journey are highly variable and unpredictable, gives rise to similar negative impacts on the environment and the economy as does traffic congestion. Moreover, being able to offer a high degree of travel-time reliability will facilitate the uptake of sustainable road transportation including future public, shared, … Read more
Offering predictable journey times is important to the uptake of sustainable road transportation including future public, shared, and on-demand mobility services and to on-time delivery of goods. To achieve such predictability, the ClearWay [1] project at TCD is exploring ‘slot-based driving’ (SBD) as a strategy for active management of roads (especially highways). SBD abstracts traffic … Read more
Poor travel-time reliability, meaning that travel times for the same journey are highly variable and unpredictable, gives rise to similar negative impacts on the environment and the economy as does traffic congestion. Moreover, being able to offer a high degree of travel-time reliability will facilitate the uptake of sustainable road transportation including future public, shared, … Read more
An ideal bus route would offer highly predictable journey times to travellers so that the same journey taken at the same time of day on different occasions would take the same amount of time. While there are many sources of variability in journey times such as traffic conditions, passenger boarding and offloading time, and the … Read more
This project will investigate the use of reinforcement learning to develop a highway journey booking system incorporating a dynamic pricing strategy to allow traffic demand to be shaped in ways that will improve traffic efficiency and enhance sustainability by reducing emissions and fuel consumption. The goal will be to evaluate the potential benefit of the … Read more