Mobile robots are considered a central element for many emerging applications such as agile manufacturing, environmental monitoring, inspection and maintainance and so forth. Within this context the use of mobile robots holds great promisises to facilitate the work of human operators in difficult, dangerous or repetitive tasks.
These kind of applications typically involve unexpected situations (due to unforseen environmental conditions or operators’ behaviors) that force the mobile robots to coordinate their actions and act as a team. Recently, a number of advancements have been made in the design of coordination procedures for these type of applications that are based on key research themes within the broad area of Artificial Intelligence such as optimization, planning and machine learning to name a few.
In this tutorial, we provide a broad overview of key issues for coordination approaches that must be deployed in mobile robot teams and discuss a selection of most promising solution techniques. Specifically, we focus on the use of task assignment approaches, graphical models and constraint processing as well as reinforcement learning methods to solve the coordination problem for robot teams. We discuss how decentralized coordination algorithms can be used within this context, presenting widely used solution techniques and discussing their merits with respect to our reference application scenarios. Finally, we conclude by highlighting open problems and possible future venues of research within this field.