A Framework for Multi-Robot Coordination (original) (raw)

In this thesis work, a complete framework for multi-robot coordination in which robots collectively execute interdependent tasks of an overall complex mission requiring diverse capabilities is proposed. Given a heterogeneous team of robots and task dependencies, the proposed framework provides a distributed, robust mechanism for assigning robots to tasks in an order that efficiently completes the mission. The approach is robust to unreliable communication and robot failures. The framework is based on the market-based approach, and therefore scalable. In order to obtain optimum allocations in noisy environments, a coalition maintenance scheme ensuring dynamic reconfiguration is introduced. Additional routines, called precautions are added in the framework for addressing different types of failures common in robot systems and solving conflicts in cases of these failures. The final solutions are close to optimal with the available resources at hand by using appropriate cost functions. The framework has been tested in simulations that include variable message loss rates and robot failures. The experiments illustrate the effectiveness of the proposed system in realistic scenarios.