Distributed active reinforcement learning for multi-agent planning and control

With the rapid development of network-connected systems, coordination and cooperation among the subsystems/agents have become increasingly important and powerful in many control and robotics applications. Benefiting from extensive interactions with unknown environments, reinforcement learning (RL) has achieved remarkable success in playing complex games and virtual robotic control. However, its applicability in real world applications remains quite limited, mainly owing to its poor sampling efficiency, i.e., a large amount of trial-and-error attempts. There is a classic dilemma of RL algorithms: should the agents maximise their reward based on their current knowledge or explore poorly understood states and actions to potentially improve future performance? Recently, RL-based path planning has received significant research attention in robotics and control community. In challenging and uncertain environments, exploring poorly understood states is of significant important for acquiring reliable knowledge and thereby ensuring mission success. To improve the sampling efficiency, this PhD project aims to develop active exploratory RL algorithms for multi-robot systems, which are in contrast to the traditional epsilon-greedy RL with random exploration mechanisms. Newly developed algorithms will be tested using benchmark applications to validate their effectiveness and improvement compared with traditional RL algorithms using simulation and experimental studies.

Eligibility

Applicants should have, or expect to achieve, at least a 2.1 honours degree or a master’s in a relevant science or engineering related discipline. Demonstrable experimental experience in robotics and control is to a great advantage.

Funding

At Manchester we offer a range of scholarships, studentships and awards at university, faculty and department level, to support both UK and overseas postgraduate researchers.

For more information, visit our funding page or search our funding database for specific scholarships, studentships and awards you may be eligible for.

Before you apply

We strongly recommend that you contact the supervisor(s) for this project before you apply. Enquiries regarding this project can be sent to Dr. Zhongguo Li ([Email Address Removed])

How to apply

To be considered for this project you’ll need to complete a formal application through our online application portal.

When applying, you’ll need to specify the full name of this project, the name of your supervisor, how you’re planning on funding your research, details of your previous study, CV, and names and contact details of two referees.

Your application will not be processed without all of the required documents submitted at the time of application, and we cannot accept responsibility for late or missed deadlines. Incomplete applications will not be considered. 

If you have any questions about making an application, please contact our admissions team by emailing [Email Address Removed].

Equality, diversity and inclusion is fundamental to the success of The University of Manchester, and is at the heart of all of our activities. We know that diversity strengthens our research community, leading to enhanced research creativity, productivity and quality, and societal and economic impact.

We actively encourage applicants from diverse career paths and backgrounds and from all sections of the community, regardless of age, disability, ethnicity, gender, gender expression, sexual orientation and transgender status.

We also support applications from those returning from a career break or other roles. We consider offering flexible study arrangements (including part-time: 50%, 60% or 80%, depending on the project/funder).