From the early days of mobile robotics, back in the 1960s, vision has always been one of the key sensors for controlling a robot. Traditionally, vision is thought of as a camera on board a single robot; this book contributes in a slightly different area: integrating the information from multiple cameras for a group of robots. Those cameras are not necessarily on board the vehicle, but are distributed around the area where the vehicles are going to move. This fact means enlarging the sensing and reasoning capabilities of the robots, but it also presents a challenge from a technical standpoint.

This monograph is full of contributions in the field of visual control of teams of robots by using multi-view geometry and monocular cameras. More specifically, the authors use and extend well-known computer vision methods and validate them through simulations and field tests (for example, homographies, trifocal tensor, and so on). The information extracted from the cameras is taken by visual control approaches to control the position of a team of mobile robots.

Another significant contribution of this monograph is related to the coordinated control setup where images taken by unmanned aerial vehicles (UAVs) are employed for controlling a team of ground robots. Novel decentralized formation stabilization methods dealing with the problem of controlling a multi-robot formation close the book.

This book is primarily written for researchers and PhD candidates with a special focus on computer vision and visual control.