This doctoral research project explores advanced control methodologies for aerial robot swarms, focusing on dynamic leader selection and robust coordination strategies that adapt to changing mission requirements and potential system failures.

Description:

Traditional leader-follower approaches in swarm robotics typically designate fixed leaders, creating potential single points of failure and limiting mission flexibility. This research will develop adaptive leadership frameworks where leader roles dynamically shift based on mission phases, environmental conditions, and system health. The project aims to create scalable, fault-tolerant control architectures that maintain swarm integrity and mission performance despite individual robot failures or changing objectives.

Objectives

• Develop algorithms for dynamic leader selection based on mission requirements, robot capabilities, and environmental factors
• Design robust control strategies that maintain swarm cohesion during leadership transitions
• Implement fault detection and isolation techniques to trigger leadership changes when failures occur
• Create scalable control architectures that perform effectively with varying swarm sizes
• Develop formal methods to verify and validate the stability and performance of the adaptive swarm system
• Evaluate the system through extensive simulation and experimental validation using physical robot platforms
• Investigate applications in search and rescue, environmental monitoring, and infrastructure inspection

Key skills/interests:

Mathematical modelling, control design, simulation, state machines.

Expected outputs:

1x PhD (Eng), 2-3x Journal/Conference paper in a leading publication.

Supervisors:

Arnold Pretorius (MechatronicSystems.Group, Dept. of Mechanical Engineering)

Eligibility:

• Full-time student working on campus
• Undergraduate GPA of at least 70%