Event-driven layered affine formation appointed-time prescribed performance control for MSVs: A fully distributed approach over mixed graphs

This study investigates the problem of event-driven layered affine formation appointed-time prescribed performance control for marine surface vehicles system (MSVs) over a mixed topology graph. Initially, a layered framework is established, delineating the flow of information across two distinct layers. At the first layer, information is transmitted from the virtual MSV to the leader MSVs. Next, at the secondary layer, this information is passed down from the leader MSVs to the follower MSVs. Furthermore, an appointed-time prescribed performance control (APPC) is developed to make the position tracking errors within the preset performance boundary with a prescribed time. Then, we introduce the minimal parameter learning (MPL) approach, which utilizes the radial basis function neural network (RBF NN) to address the lumped model uncertainties for MSVs by updating the norm of the weight matrix. This streamlines the control design process and reduces the computational load, offering a more efficient alternative to conventional RBF NN. In addition, a relative threshold-based event-driven mechanism is introduced to significantly decrease the network bandwidth usage. The controllers are subsequently constructed for MSVs in both layers to achieve the target formation tracking objective using the previously mentioned methods. Finally, two distinct numerical simulation results confirm the effectiveness and advantages of the proposed controller.