Postgraduate research project

Multi-agent reinforcement learning for safe navigation of multiple robots

Funding
Competition funded View fees and funding
Type of degree
Doctor of Philosophy
Entry requirements
2:1 honours degree View full entry requirements
Faculty graduate school
Faculty of Engineering and Physical Sciences
Closing date

About the project

As robot swarms increasingly operate alongside people, robots must do more than avoid collisions. They must behave in ways that people find safe, predictable and comfortable. This project investigates how multiple robots can coordinate their collective goals while adapting their behaviour to individual humans, preferences and social expectations in shared spaces.

Robot swarms could support people in homes, workplaces, hospitals and public spaces, but operating safely around humans requires more than collision avoidance. Robots must balance their own objectives with the safety, comfort, expectations and trust of the people sharing their environment.

This project investigates human-centred coordination for multiple robots operating in shared spaces. The research explores how multiple robots can pursue collective tasks while adapting their behaviour to the presence of different people. Rather than treating humans simply as obstacles, the project investigates how robots can account for individual preferences, comfort zones, social expectations and changing human behaviour when deciding where and how to move.

Building on existing work using decentralised multi-agent reinforcement learning for navigation around multiple humans, the project explores questions such as: 

  • how should a swarm trade off task efficiency against human comfort?
  • how can robots learn individual differences in acceptable proximity and behaviour?
  • how should robots coordinate when accommodating one person's preferences affects the swarm's task?
  • how can a swarm communicate its intentions so that people understand and trust its behaviour?

The research may combine multi-agent reinforcement learning, human-robot interaction, robot navigation and experimental studies with humans, using simulation and physical robot platforms. You'll have opportunities to develop and evaluate novel algorithms and interaction strategies for safe, socially acceptable and trustworthy robot swarms operating in real-world environments. You'll have access to the robotic platforms available in SooratiLab including robot dogs, several ground rovers, quadcopters, robotic arms, etc.  

The School of Electronics and Computer Science is committed to promoting equality, diversity inclusivity as demonstrated by our Athena SWAN award. We welcome all applicants regardless of their gender, ethnicity, disability, sexual orientation or age, and will give full consideration to applicants seeking flexible working patterns and those who have taken a career break. The University has a generous maternity policy, onsite childcare facilities, and offers a range of benefits to help ensure employees’ well-being and work-life balance. The University of Southampton is committed to sustainability and has been awarded the Platinum EcoAward.