Control of Launch and Recovery in Enhanced Sea-States: Part of the Launch and Recovery Co-Creation Initiative

目前，许多海上作业，例如从小型飞行器、有人驾驶和无人驾驶航空器和潜水器的母船上发射和回收，只能在足够平静的海况下安全地尝试。作为示例，小型船舶从母船的L&R通常涉及在经由起重机/提升机构的两个船舶的主要物理连接之前，两个船舶靠近地一起移动（通过船首线连接）。在许多情况下，整个任务的波浪临界高风险要素，即小型船舶与母船的连接和随后的提升，仅持续几十秒。超过这个时间会增加手术风险。一旦两个飞行器物理连接，操作员就开始启动提升过程。在这种情况下，即使是对船舶运动的静止期的短期预测，也具有相当大的操作价值，并且可以允许在目前被认为不合适的条件下安全地取消L&R。这种增强的L&R能力对现代海军非常有吸引力。在这个项目中，研究的目的是开发一种新的方法来预测一个合适的时刻，在该时刻启动一个L&R操作，连同一个置信度测量（作为建议提供给人类操作员），然后控制执行随后的电梯操作一旦启动，使用一种新的形式的模型预测控制（MPC）。主要的项目交付成果是：（一）一个原型决策支持系统（DSS），在软件模拟器中运行，在有限的时间范围内对回收过程的各个方面提供持续更新的短期预测模拟;（二）实际物理提升过程的控制器。这两个要素将利用由确定性海浪预测（DSWP）系统的波浪预测以及历史和实时船舶运动传感器数据驱动的水动力船舶运动预测模型。DSS最初将在小艇接近母船并拿起一根船首索（低风险活动）时启动，但尚未连接到提升机构。该研究将假设存在于一个通用的绞车/起重机起重系统与一个单一的电缆的母舰。缆索张力是一个关键的控制量，最大可用提升力是系统的主要规格参数。安保部将：（i）确定在预测的静止期内连接提升绳并开始提升的适当时刻;（ii）为该特定模拟提升的安全/成功提供置信度测量。通过拍摄两艘船当前状态的快照（用作初始条件）以及母船运动的短期预测来确定连接和提升的适当时间，以模拟是否有可能使用MPC控制器成功回收小型船舶。然后将向操作员呈现包括置信度度量的当前建议摘要。如果作为该建议连接和提升的结果未被发起，则该过程使用新的当前数据的快照来重复。此循环将继续，直到操作员决定接合提升机构（或回收中止）。当连接/提升实际上被启动时，物理提升阶段然后将采用在模拟中使用的相同MPC控制器，利用母船的运动的预测、两个船舶的实际实时测量的运动以及小型船舶在悬离水面时的自由体模型。

Currently many marine operations, such as the Launch and Recovery (L&R) from a mother ship of small craft, manned and unmanned air vehicles and submersibles, can only be attempted safely in sufficiently calm sea-states. As an example, the L&R of a small craft from a mothership typically involves the two vessels moving together in proximity (linked by a bow-line) before the main physical connection of the two via a crane/hoist mechanism. In many cases the wave-critical high risk elements of the overall task, i.e. the connection and subsequent hoist of the small craft to the parent vessel, only last for a few tens of seconds. Taking longer than this increases the operation at risk. Once the two craft are physically connected the operator is committed to initiate the hoisting process. In this context even the short term prediction of quiescent periods of vessel motion resulting from lower than average wave activity in otherwise large sea states, has considerable operational value and may allow L&R to be untaken safely in conditions which would currently be deemed unsuitable. Such enhanced L&R capabilities are very attractive to modern navies. In this project the research aim is to develop a novel approach to predicting a suitable time instant at which to initiate an L&R operation, together with a confidence measure (provided as advice to a human operator), and then to control the execution of the subsequent lift operation once initiated, using a novel form of Model Predictive Control (MPC). The key project deliverables are: (i) a prototype decision support system (DSS), running within a software simulator, which provides continuously updated short term predictive simulations over a finite-time horizon of all aspects of the recovery process; (ii) a controller for the actual physical hoist process. These two elements will exploit hydrodynamic vessel motion prediction models driven by wave predictions from a Deterministic Sea Wave Prediction (DSWP) system, and historical and real-time vessel motion sensor data. The DSS will initially be engaged as the small craft approaches the mothership and picks up a bow-line (a low risk activity), but is not yet attached to the hoist mechanism. The research will assume the presence on the mothership of a generic winch/crane lifting system with a single cable. The cable tension is a key controlled quantity and the maximum lifting force available is a major system specification parameter. The DSS will: (i) identify an appropriate moment to attach the hoist line and initiate hoisting during predicted quiescent periods; (ii) provide a confidence measure for the safety/success of that specific simulated lift. An appropriate time to attach and hoist will be identified by taking a snap-shot of the current state of both vessels (to use as initial conditions) together with short term predictions of the movement of the mothership to simulate whether it is possible to successfully recover the small craft using the MPC controller. The operator will then be presented with a current advice summary including confidence metrics. If as a result of this advice connection and hoisting is not initiated, the process repeats using a snapshot of the new current data. This cycle continues until the operator decides to engage the hoist (or the recovery is aborted). When connection/hoisting is actually initiated, the physical lifting phase will then employ the same MPC controller used in the simulation, exploiting predictions of the motion of the mothership, the actual real-time measured motions of both craft and a free body model of the small craft when suspended clear of the water.