Autosea
Sensor fusion and collision avoidance for autonomous surface vehicles - Master thesis proposals
Collision avoidance projects
- Risk based collision avoidance using Reachable sets and Bayesian methods
- Collision avoidance for autonomous surface vehicles focusing on COLREGS
- Collision avoidance for autonomous surface vehicles using the branching course MPC algorithm
- Uncertainty management in scenario-based MPC for collision avoidance
- Uncertainty management in scenario-based MPC for collision avoidance
- Obstacle movement prediction for ASV collision avoidance
- Collision avoidance system for ships utilizing other vessels' intentions
Sensor fusion projects
- Object detection and classification for autonomous navigation systems
- Camera Positioning for Unmanned Ships
- Active-Passive Sensor Fusion
- Detection of ships in infrared data
- Near-shore target tracking with map uncertainty
- Detection of ships in infrared data
- Heading accuracy in single-target tracking
- Sensor fusion using radar and AIS
Miscellaneous projects
- Requirements to object detection of autonomous navigation systems based on ship maneuverability and speed
- Automatic model identification and kinematic control of high-speed autonomous surface vehicles
- ReVolt – Demonstration platform
- A ship motion control concept inherently satisfying actuator constraints
- Adaptive motion control for ships with cascaded nonlinear feedback control
- HIL-testing for maritime collision avoidance systems
- Dynamic positioning for ASV
This page presents project- and master thesis proposals for the Autosea project. The project has gathered considerable ammounts of data in the Trondheimdsfjord during the winter of 2015/2016 and the spring 2017 together with Maritime Robotics. Many of the results from the suggested theses can be verified with the gathered data, or tested in new experiments. Matlab and Python are the main tools used for development and simulation. Some project software is developed using ROS
General information about the project can be found at AMOS' website.