Autonomous Mobile Robots and Multi-Robot Systems

10Collective Motion with Shared Environment Map

Eugene Kagan and Irad Ben‐Gal

This chapter considers motion of the group of robots following common environment map, which is defined by the terrain or certain potential field, and stresses the differences between collective and swarm dynamics. In addition, the chapter considers the motion with and without leader. The discourse is illustrated by probabilistic search algorithms and the methods of obstacle and collision avoidance based on attraction/repulsion potentials.

10.1 Collective Motion with Shared Information

In the previous chapter, we considered general principles of swarm dynamics and indicated different types of communication and information transfer that are require for collective activity of the agents. In the simplest case, the agents communicate according to certain communication protocol and have complete information about the states of the environment and the actions of each agent in the group. In practice, such a situation is achieved by the use of central unit, which obtains information from each agent and shares it with the other agents. In the other case, the agents do not use central unit and transfer information about their local environment to other agents in the group. Then, motion planning is conducted using the environmental map and corresponding information, which are computed by each agent itself. These two situations are illustrated in Figure 10.1.

Figure 10.1 Sharing information: (a) using ...