A novel strategy to solve communication constraints for formation control of multi-AUVs

Interest in terms of research on formation control of multiple autonomous underwater vehicles (multi-AUVs) has gained considerable attention in recent years [1–5], owing to its extensive applications in deep sea inspection, ocean sampling and localization. The formation control objective cannot be realized without information (i.e., position and velocity) exchange between AUVs. Owing to the particularity of underwater environments, acoustic communication is the most effective data transmission mode. However, this mode still has many drawbacks, such as long propagation delays, high transmission losses, packet drops, and limited bandwidth. Reducing the amount of data is an effective way of decreasing the impact of such defects. Generally speaking, there are two main strategies to decrease the amount of data in general communication systems. First is to reduce the data in each packet, and the second is to reduce the amount of data packets. A reasonable way to approach the first strategy is removing the transmission of speed information between AUVs. Owing to environmental disturbances and/or technology limitation, the accurate velocity measurement of AUVs is more difficult than position measurement. To obtain the velocity of AUVs, the observer is a common way, which is used in many studies [4,5]. In [4], a high-gain observer-based cooperative path following control approach was developed without measuring the velocity of vehicles. In [5], a global finite-time convergent observer was proposed to reconstruct vehicle velocity. Neither the linear observer in [4] nor the nonlinear observer in [5] can guarantee convergence within the setting time. An effective way to approach the second strategy is to reduce the communication frequency between different subsystems. To this end, some new communication mechanisms have been proposed in [6, 7]. Among these methods, eventtriggered communication mechanism is of significant interest. The key idea of the event-triggered theory is to sample and update the controller only when certain conditions are violated. If the threshold is set properly, this scheme can reduce the required computation cost while maintaining satisfactory control performance. With regard to the above observations, this study proposes a novel strategy to solve the communication constraint problem. This new strategy combines the eventtriggered transmission scheme with the velocity observer method. This strategy is capable of not only decreasing the data in each packet but also can reducing the amount of data packets. Kinematics and dynamics model of AUV. The kinematics and kinetics of the AUV can be described as [8]