On Three-dimensional Formation Control with Mismatched Coordinates
Date
2017-07-17
Authors
Meng, Ziyang
Anderson, Brian
Hirche, Sandra
Journal Title
Journal ISSN
Volume Title
Publisher
IEEE
Abstract
A typical formation shape control problem involves point agents sensing relative positions, i.e. orientations and distances, of their neighbors and then moving so that these relative positions achieve some prespecified values. Such a procedure, requiring as it does sensing of orientations, implicitly presupposes that all agents have a shared understanding of the common orientations. On the other hand, there may be biases in sensors, variations in the earth's magnetic field interfering with compass-based sensing, or drift in inertial sensors, with the result that orientations are inconsistently measured or measured with error. In this paper, we investigate the formation control problem with mismatched coordinates in the three-dimensional space, considering the consequences of this error. First, the situation of a two agent formation is first considered. We show that the agents converge to a fixed, but distorted formation exponentially fast. In contrast to the matched case, the formation is not asymptotically stationary, but rather instead translates with a certain constant velocity depending on the mismatches. The formation distortion between the actual one and the desired one is obtained, as well as the steady state velocity of the formation for small mismatch orientations. The case of agents with double integrator dynamics is then considered and similar phenomena are observed. Based on the results, an estimation algorithm is given to obtain the mismatch rotation matrix, which allows a compensation algorithm to be proposed such that the desired formation is achieved with zero steady-state velocity for the formation as a whole. The case of n-agent formations is finally considered, first with a star graph and then with a general graph. Simulations are provided to validate the theoretical results.
Description
Keywords
Autonomous vehicles, control system analysis, measurement errors
Citation
Collections
Source
IEEE TRANSACTIONS ON CONTROL OF NETWORK SYSTEMS
Type
Journal article
Book Title
Entity type
Access Statement
Open Access
License Rights
Restricted until
Downloads
File
Description