System Identification and Control of a Micro Autonomous Surface Vessel
(Room Innovation Hub-- Booth 2727)
04 May 16
10:00 AM
-
5:30 PM
Tracks:
Academic, Defense, Maritime, Research and Development
The presentation includes vessel schematics for the monohull testing vessel and the final trimaran vessel. The hardware list is included along with the decision matrices and justification for each part. Both functionality and availability were considered when deciding on which parts to use for the vessel. The dynamics of the ASV were determined by giving a series of commands and monitoring how the vessel responded. At various speeds, the vessel was commanded to do a hard right or hard left turn by using differential drive. The particular measurements of interest were the acceleration in the x, y, and z directions as well as the roll, pitch, and yaw. From the data gathered, there were options that could be used to create a model of the vessel. By knowing the general form of the linearized dynamic equations of the system, it is possible to extract data from the results to generate a system model that shows good agreement with the actual system using simple mathematical methods. Another, more complex and yet potentially more accurate model-based approach involves adaptive control. Adaptive control works by maintaining a dynamically adjusted model of the system behavior over time. The difference at each time step between the expected outcome and the actual output of the system is used to dynamically adjust the parameters of the model. The adjustment mechanism will ideally change the parameters of the model until the output converges to the desired output. This method of control can be done directly through MATLAB/Simulink using the TRAC algorithm. The final models based on the originally obtained data are included. To ensure that the model fits the actual dynamics of the vessel, it should be tested against a new set of data. Once again, extreme commands were given to the ASV given while monitoring the acceleration and orientation in space. The model was compared to the new set of data. The dynamics of the trimaran were expected to be similar to the mono hull vessel in the sense that both vessels were of similar size and used differential drive. The same sort of model development was used on the trimaran as the monohull vessel. After developing a feedback controller, the ASV was tested in controlled waters and then moved to the Chesapeake for a final test run. The final result is a small ASV that could successfully navigate for up to seven hours in open water and return to its starting location. Micro ASVs show a strong potential for both scientific and military. Due to their small size, they are easy to launch and recover and have a low cost. They could be used in a wide range of missions such as gathering data to study the environment to surveillance on adversaries.