<p>The use of autonomously underwater vehicles (AUVs) has a great potential in scientific mission involving underwater exploration. However a major drawback with todays AUV missions is the launch and recovery process which are usually performed manually from a manned supply ship. These manned ships have a huge daily operation cost, and because AUVs can have operation times up to 70 hours these missions become extremely costly. Since the combination of an AUV together with a manned mothership is very costly the use of AUVs are very restricted. A solution here is to replace the manned mothership with an unmanned vehicle such as a unmanned surface vehicle (USV). This will reduce the cost of AUV mission drastically and therefore increase the use of AUVs on scientific missions. This motivates the need for an AUV-USV docking method which is one of the two docking scenarios treated in this master thesis. Another docking method treated here is the possibility to dock a USV together with a manned mothership without human interference. A docking method that removes the human intervention will make the USV completely unmanned, since USVs today are manually docked together with a mothership or driven back to shore by a remote control. To achieve an understanding of the field, a summary of the most relevant findings in todays literature are given. This includes the possibility to autonomously dock together an AUV with another vehicle or installation, and other related fields such as spacecraft docking and aerial refuelling. The main findings involving AUV docking, ranges from a simple fuzzy logic procedure to more advanced methods involving trajectory planning and potential field guidance. Since no extensive previous work exist on general USV docking, a short introduction is given to the most related fields, such as spacecraft docking and aerial refuelling. During air refuelling two methods are summarised which includes racetrack pattern or waypoint paths, where the receiver aircraft has two different ways of rendezvousing with the tanker, namely point parallel- or route-rendezvous. In both docking scenarios treated here, rendezvous guidance is developed since the vehicles are assumed underactuated. The docking procedure is divided into two stages, a homing stage and a docking stage. In the homing stage only rough guidance is needed which is not the case during docking stage where requirements are much tighter on positioning to avoid collisions. In the AUV to USV homing stage the USV does all the work, but during docking stage the AUV has full responsibility, since the USV only traverses along a straight path. The USV's path is here orientated against the wind direction to minimise the sideslip effect caused due to weather disturbances. Once the USV has converged to a straight path the AUV proceeds to docking from behind the USV to finalise docking. For the USV to mothership docking scenario, the USV has the full responsibility during the whole docking procedure. Here the USV is underactuated, and therefore the mothership will be in motion and only has to avoid sudden manoeuvres. In the homing stage the USV will manoeuvre towards a point given on the line of sight vector between the two vehicles. Once the USV reaches this point it will steer along a circle around the mothership to avoid collisions and to position itself in clear sight of the docking point. With clear sight achieved the USV will use its forward motion to converge sideways towards the docking point, such that docking can be completed. Finally, simulations are carried out to verify the behaviour of the developed guidance laws. During these simulations two 3DOFs underactuated USV models are being used, where both vehicles only has controllability over surge speed and yaw rate. In both docking scenarios the whole docking procedure is analysed including homing and docking stage. The simulation results shows a proper docking with a satisfying approach in both scenarios. Also the mothership's velocity is examined to understand the USV's sideway approach towards the mothership.</p>
| Identifer | oai:union.ndltd.org:UPSALLA/oai:DiVA.org:ntnu-10039 |
| Date | January 2010 |
| Creators | Loberg, Jon-Erik |
| Publisher | Norwegian University of Science and Technology, Department of Engineering Cybernetics, Institutt for teknisk kybernetikk |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, text |
Page generated in 0.0023 seconds