Temporal Coordination among Two Vision-Guided Vehicles: A Nonlinear Dynamical Systems Approach

The field of multiple autonomous robots cooperating is emerging as a key technology in mobile robots and is currently under intense effort. The use of multi-robots synchronized, coordinated or cooperating in production processes where there is a high requirement on flexibility and manoeuvrability is highly desirable. This is an option to be considered in complex and integrated production processes including assembling, transporting, painting and welding tasks. Broadly, the applied general approaches for controlling and coordinating the movement of several robots that cooperatively perform a task illustrate the major trade-off in the control and coordination of multi-robots: between precision and feasibility and between the necessity of global information and communication capacity. Further, multiple robot systems working in external synchronization, e.g master-slave and coordinated schemes or mutual synchronization, e.g. cooperative schemes, imply the design of suitable controllers to achieve the required synchronous motion. The work presented in this paper, combines insights of computer vision, dynamical systems theory, computational neuroscience and robotics. We aim at generating online flexible timed behavior stably adapted to changing online visual, infrared and proprioceptive sensory information, such that different entities may achieve autonomous timed and flexible cooperative/coordinated behavior. As a first attempt, we do not take into account communication issues. We apply an attractor based dynamics as recent studies have shown that this theory helps synchronize systems and reduces the computational requirements for determining identical movement parameters across different coupled entities. The inherent advantages from an engineering viewpoint are huge, since the control system is released from the task of recalculating the movement parameters of the different entities. The main motivation is that once solutions for this problem are found, they can be applied in search and rescue operations, landing removal, remote terrain and space exploration, and also to the control of satellites and unmanned aerial vehicles. In this domain, the achievement of robots able to exhibit intelligent and flexible cooperative behaviour is a first issue. The approach is demonstrated in the cooperation among two vision-guided mobile robots such that they are able to reach a visually acquired goal, while avoiding obstacles, without O pe n A cc es s D at ab as e w w w .ite ch on lin e. co m