Data-driven machine learning techniques are popularly used in the field of robotics to inform autonomous decision-making and to perform control or path-planning. Supervised learning and reinforcement learning have been shown to be particularly amenable to canonical tasks that are integral to robotics applications. However, these techniques rely on data in the form of action-label (supervised), action-value (regression) or action-reward (RL) pairs, where the action is a path (or some other) execution by a real robot. e.g.
EPSRC-Eligible and EU applicants
Below, we list PhD topics for EPSRC-Eligible and EU applicants only. For overseas applicants see here.
To be EPSRC-Eligible for a full award, an applicant must have no restrictions on how long they can stay in the UK and have been ordinarily resident in the UK for at least 3 years prior to the start of the studentship (with some further constraint regarding residence for education).
Subsea inspection of structures is now commercial and the next frontier in subsea robotics is the safe physical interactions with underwater structures. This requires the development of new control algorithms with force compliance which can take into account external disturbances. There is also a need to work across a variety of control models, from full teleoperation across high bandwidth data to shared autonomy (our goal) across low and intermittent connection to enable shore-based control of remote platforms.
The Internet-of-Robotic-Things (IoRT) brings together autonomous robotic systems with the Internet of Things (IoT) vision of sensors and smart objects pervasively embedded in everyday environments [1-4]. This merge can enable novel applications in almost every sector where cooperation between robots and IoT technology can be imagined. Early signs of this convergence are in network robot systems , robot ecologies , or in approaches such as cloud robotics .
Curiosity guides humans to learn efficiently. It incentivizes us to spend more energy and time examining new, unexpected things, and to disregard those we fully understand already, to make our learning more efficient. Much of the vision learning that is done today is passive: learning systems are exposed to large amounts of training data, and learn from each sample multiple times, regardless of their current ability to recognize them at the time. This makes the process slow, specially given the increasingly large number of samples on datasets.
Human action recognition is a fundamental problem that underlies many applications in robotics, including interaction, home care, collaboration, etc. The actions that can be recognized by robots or computers today are often coarse and simplistic, in the sense that they are very different from each other; for example eating vs playing piano, or sitting vs standing. Both datasets and technology tend to be broad and crude. As human-robot-interaction becomes more natural we require more sophisticated technology for perceiving humans.
Vision is a key ability to humans as well as to robots to understand and extract information from real world and is crucial to safely interact with our environments. In contrast to human perception, state-of-the-art machine vision methods require millions of images and their manual labels to learn each visual task. This project focuses on designing machine learning and computer vision techniques that can help robots to learn multiple tasks from limited labelled data.
There are two main directions for potential projects:
Natural Language Generation (NLG) is the task of translating machine-readable representations and data into human language, and thus vital for accountability in safe human-machine collaboration. Neural Network architectures for NLG are promising since they able to capture linguistic knowledge through latent representations using raw input data, and hence have the benefit of simplifying the design of systems by avoiding costly manual engineering of features, with the potential of more easily scaling to new data and domains.
The training phase in Deep Learning is very compute and data intensive and, therefore, the efficiency of the training phase typically restricts the quality of results that can be achieved within a given time frame.
Many learning algorithms are dominated by the speed in which data can be brought to the CPU, i/e., by the memory bandwidth of the executing hardware. Consequently, techniques that are based on reduced precision number representations have been shown to produce faster results without a significant loss in the quality of results.
How can we ensure that robots behave in a way as desired by humans? To ensure that the execution of complex actions leads to the desired behavior of the robot, one needs to specify the required properties in a formal way, and then verify that these requirements are met by any execution of the program.
The aim of this project is to develop a stringent approach to automatic programming of control systems based on an intermediate implicitly learned representation of the control task by a neural network. While deep neural networks (DNN) have been shown to be capable of solving control problems effectively based on learning from demonstration and reinforcement learning, the resulting representations are computational complex and lack immediate inspectability.