IEEE - Aerospace and Electronic Systems - August 2022 - 21

Image licensed by Ingram Publishing
STRUCTURE OF THE SYSTEM
According to Figure 2, the stand-on object
S and C
LITERATURE
PS and the
give-way object PG consist of the UAVs PS and PG,
which are independently controlled by local two-degreesof-freedom
controllers C
G. They generate the control
inputs uSðtÞ and uGðtÞ and make the UAVs follow
their trajectories generated by the corresponding control
units AS and AG. These units are able to communicate
with one another over the communication network. They
use the locally measured positions pSðtÞ and pGðtÞ to generate
the matrices WSðtÞ and WGðtÞ, which consist of the
trajectories and their first four derivatives. The current
time delays of the transmission of data packets are estimated
by the delay estimators DS and DG. In the figure
the solid arrows indicate a continuous signal transmission.
The dashed arrows represent an event-based signal transfer,
where signals are only sent if a threshold is violated.
For modeling, the communication network is subdivided
into the receiver parts indicated by RS and RG and
the transmitter parts TS and TG for the stand-on object and
the give-way object, respectively. The transmission of the
data is represented by the Ricean fading model with the
function fRðRsðdð~tiÞÞjPrðdð~tiÞÞ; KÞ. The signal propagation
is modeled by Ricean fading as in the free airspace
there is a strong line-of-sight path between the UAVs with
only small reflected and diffracted components [2], [3]. It
characterizes the small-scale fading. In [4] an overview
over different UAV channels is provided that includes the
air-to-air communication as considered in this article as
well as air-to-ground and ground-to-ground connections.
The cooperative control of autonomous UAVs has
received increasing interest in recent years. Using a group
of UAVs allows the execution of more complex tasks
compared to the use of a single UAV. Furthermore, reliability
and flexibility are increased [5], [6]. Application
fields include search and rescue operations [7], [8], geographic
mapping [9], or the surveillance of large
areas [10], [11]. In order to fulfill these tasks, several challenges
including the provision of suitable communication
links between the objects as well as the collision avoidance
need to be addressed [5]. In this article, the problem
of collision avoidance is investigated, where necessary
information (e.g., positions of the objects) are only communicated
over a network. As these communication links
are assumed to be uncertain, methods should be developed
to cope with these communication uncertainties. The scenario
to be investigated is similar to search and rescue
missions, where the UAVs should not collide and should
not exceed a maximum distance to one another to enable a
real-time communication (e.g., a video stream of the current
situation) between them and to a base station [8].
The proposed approach combines methods from communication
technology and control theory to satisfy the
control aims in the presence of varying time delays and
packet losses. In communication technology many different
approaches have been developed for the modeling of
wireless channels [12]. For UAV applications a two-state
Figure 1.
Scenario to be investigated.
Figure 2.
Structure of the networked control system.
AUGUST 2022
IEEE A&E SYSTEMS MAGAZINE
21
IEEE - Aerospace and Electronic Systems - August 2022

Table of Contents for the Digital Edition of IEEE - Aerospace and Electronic Systems - August 2022
