Show simple item record

dc.contributor.advisorPettersen, Kristin Ytterstad
dc.contributor.advisorEngelhardtsen, Øystein
dc.contributor.authorStenersen, Thomas
dc.date.created2015-06-16
dc.date.issued2015
dc.identifierntnudaim:12747
dc.identifier.urihttp://hdl.handle.net/11250/2352498
dc.description.abstractThere has been a rapid growth in autonomous technology in several fields the past decade. One of these is autonomous navigation, where the DARPA Grand Challenges has been a major research catalyst. Today, the Centre of Excellence "Centre for Autonomous Marine Operations and Systems (AMOS)" in cooperation with the maritime industry seeks to develop autonomous technology for marine applications. A robust collision avoidance system is crucial for an Autonomous Surface Vessel (ASV). In order to operate at sea, near other traffic, it also needs to adhere to the "rules of the road", the International Regulations for Avoiding Collisions at Sea (COLREGs). In this thesis, a COLREGs compliant guidance, navigation and control (GNC) system has been developed with the Velocity Obstacle (VO) method as the basis for collision avoidance. To validate the performance of the GNC system, a general nonlinear 3-DOF surface vessel simulator has been implemented. It features simple models for slow and fast-varying disturbances (e.g. waves, wind and current). The system has been implemented in the Robotic Operating System (ROS) framework. Three main scenarios has been designed to test the guidance system in all main COLREGs scenarios: overtaking, head-on and crossing. Each scenario requires multiple COLREGs compliant maneuvers. Several special cases are also examined and discussed. The VO method performs very well in the simulated scenarios and its avoidance maneuvers complies with all main COLREGs requirements. A thorough discussion highlights both advantages and disadvantages with method and suggests actions to mitigate less ideal behavior experienced in some special cases. The thesis also features a rich discussion on further development and improvements to the system.
dc.languageeng
dc.publisherNTNU
dc.subjectKybernetikk og robotikk
dc.titleGuidance System for Autonomous Surface Vehicles
dc.typeMaster thesis
dc.source.pagenumber123


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record