Multi-vessel sea state estimation utilising swarm shepherding

Sea State Estimation (SSE) remains critical to maritime operations worldwide. Presently, data on sea states is primarily collected by moored wave buoys in a relatively limited number of locations. Recently, research into the use of Uninhabited Surface Vessels (USVs), acting as an on-site wave buoy, has emerged. By measuring the response of the USV to wave motions and implementing a wave buoy analogy model, directional wave properties can be estimated. This study investigates the benefits of utilising an intelligent swarm of USVs (greater than two), with each member acting as a wave buoy, to estimate an aggregated sea state in real-time. Swarm intelligence can be defined as an interaction between relatively simple agents leading to the emergence of complex behaviours. It is hypothesised that a USV swarm can collaborate to enhance sea state characterisation. The challenge of swarm guidance is addressed by introducing a shepherd which guides the swarm of USVs with the intent of achieving the desired objective estimation. This method of control mimics the way a sheepdog is able to herd sheep to a target destination. A heuristic is presented which describes how a series of influential force vectors can generate swarm behaviours with the aid of a shepherd, and be used to distribute a network for the purpose of estimating the properties of a complex sea environment. The ability to estimate wave properties, with a higher degree of confidence, in-situ in any geographic region would allow maritime vessels to enhance operability, improve safety, and extend service life.