Motion and Task Planning for Heterogeneous Multi-Agent Robotic Systems

Heterogeneous Multi-Agent Robotic Systems (HMARSs) have received tremendous attention from academia and industry due to the unique advantages of tackling complex tasks more efficiently, by subdividing them into smaller tasks and executing collaborative missions. One of the most significant challenges in robotics lies in the area of multi-agent motion and task planning. The main aim of the INSPECT project is the development of efficient solutions for the problem of motion and task planning, aiming to reduce the computational complexity that stems from the HMARSs allowing computations in real-time under temporal constraints. The high-level specification will be given by a Temporal Logic (TL) formula to the system and low-level controllers will be designed to satisfy the given specification, guaranteeing the execution of collaborative tasks. To this end, INSPECT focuses on building methodologies in a decentralised fashion for heterogeneous multi-robot systems. The developed algorithms will target to address real-world problems in the robotics area including optimal inspection tasks, surveillance and monitoring, coverage, exploration and mapping by providing efficient and novel solutions. INSPECT aims to develop novel solutions based on HMARS, that will be highly versatile and scalable based on application needs, offers direct usability functions, and cost-effective. INSPECT heterogeneous multi-robot system consists of:

1. Multiple Unmanned Aerial Vehicles (UAVs)
2. Multiple Unmanned Ground Vehicles (UGVs)

The main goal of the project is the advancement of the state-of-the-art motion and task-planning algorithms for HMARSs by focusing on four main scientific innovations:

• i) to develop novel cooperative decentralised motion and task planning algorithms for HMARSs,
• ii) to reduce the computational complexity that stems from the multi-robot system,
• iii) to deal with communication, mobility, and energy constraints,
• iv) to test and validate the developed algorithms in real-world applications.