Accuracy in Parallel Dynamic Task Allocation for Multi-Robot Systems Under Fuzzy Environment

Making the right decision is an essential requirement for the task allocation process in multi-robot systems functioning in dynamic environments. Robots are often forced to make these decisions individually without any communication between them. It may be due to reasons related to uncertainty in environments or related to tasks security, such as military applications. However, robot decisions must be precise in order to increase their efficiency to perform complex tasks. This paper presents a model in which a criterion of accuracy in tasks allocation process in an uncertain environment is defined. In order to increase this precision in such environments, the robots will formulate their observations in terms of the fuzzy linguistic variables. These variables are used by a fuzzy inference system to determine a utility value of a task that most effectively increases accuracy in task allocation. Simulation results on a complex task of goods transportation by mobile robots are presented to demonstrate the effectiveness of this model.