Students' Theses
[190906] |
Title: Nonlinear Model Predictive Control for Swarms of \textgreek{m}AUV. |
Written by: {Pulido Guerrero}, Santiago |
in: (2020). |
Volume: Number: |
on pages: |
Chapter: |
Editor: |
Publisher: Institute of Control Systems: |
Series: |
Address: Hamburg, Germany |
Edition: |
ISBN: |
how published: |
Organization: |
School: {Hamburg University of Technology} |
Institution: |
Type: Project Work |
DOI: |
URL: |
ARXIVID: |
PMID: |
Note: Hespe, Christian and Saadabadi, Hamideh and Werner, Herbert
Abstract: This report presents the design of two quasi-linear parameter varying model predictive control schemes (quasi-LPV MPC) for tracking control of a nonholonomic underwater vehicle. First, a nonlinear holonomic model of an underwater vehicle is constructed and presented. Then considering the nonholonomic constraints of a real micro submarine (Hippocampus), the presented model is modified, and the system's controllability is analyzed using the kinematics of the vehicle and a Lie Bracket iteration approach. After that, the design of two quasi-LPV MPC schemes is explained and summarized in a pseudoalgorithm. Furthermore, to test the tracking performance of the proposed schemes, a helical trajectory is employed. Then the results of the simulations are compared with the tracking performance of a conventional PD-PD geometrical control. Finally, the formation control of three agents using a distributed local control strategy is presented and evaluated. Also, a novel consensus based on a predicted trajectory is considered and compared with the conventional consensus protocol.