Multi-robot team formation control in the GUARDIANS project

Abstract

Abstract. PurposeThe GUARDIANS multi-robot team is to be deployed in a large warehouse in smoke. The team is to assist firefighters search the warehouse in the event or danger of a fire. The large dimensions of the environment together with development of smoke which drastically reduces visibility, represent major challenges for search and rescue operations. The GUARDIANS robots guide and accompany the firefighters on site whilst indicating possible obstacles and the locations of danger and maintaining communications links. Design/methodology/approach In order to fulfill the aforementioned tasks the robots need to exhibit certain behaviours. Among the basic behaviours are capabilities to stay together as a group, that is, generate a formation and navigate while keeping this formation. The control model used to generate these behaviours is based on the so-called social potential field framework, which we adapt to the specific tasks required for the GUARDIANS scenario. All tasks can be achieved without central control, and some of the behaviours can be performed without explicit communication between the robots. Findings The GUARDIANS environment requires flexible formations of the robot team: the formation has to adapt itself to the circumstances. Thus the application has forced us to redefine the concept of a formation. Using the graph-theoretic terminology, we can say that a formation may be stretched out as a path or be compact as a star or wheel. We have implemented the developed behaviours in simulation environments as well as on real ERA-MOBI robots commonly referred to as Erratics. We discuss advantages and shortcomings of our model, based on the simulations as well as on the implementation with a team of Erratics. Originality/value This paper discusses the concept of a robot formation in the context of a real world application of a robot team (swarm).

Using Agents in Virtual Environments to Assist Controllers to Manage Multiple Assets

“…The question of which data should be transferred to the virtual agent is an open one: this agent must have sufficient data reason accurately about its physical counterpart, but will not necessarily require all the raw sensor data. There is also the possibility that the virtual agent may have incomplete knowledge about its physical counterpart, and it must be able to reason appropriately in this situation 5 .…”

“…While it is relatively straightforward for a single human to remotely control a single robot, teams of robots can be used more complex and efficient searches. It has still proved necessary for individual human operators to provide both the intelligence and coordination [5].…”

Experiments in cooperative human multi-robot navigation

“…Condition 2 states that the group is in a flexible formation if there exists a connected graph, which is not the case in the torn formation situation. Initial considerations definitions about the aforementioned flexible formation are introduced in our previous work [9][10][11].…”

Formation control using GQ(λ) reinforcement learning

“…For example, the formation control of robotic wheelchairs which are accompanied by human caregivers [1], or the multi-robot formation control in shared rescue missions [2].…”