Modelling and Visualization of Robot Coalition Interaction through

Smart Space and Blockchain

Alexander Smirnov and Nikolay Teslya

SPIIRAS, 39 14th Line, St. Petersburg, Russia

Keywords: Coalition, Visualization, Modelling, Robot, Intelligent Agent, Blockchain, Smart Space.

Abstract: Nowadays the study of interaction models of intelligent agents is one of the main directions in the field of

joint task solving. It includes studies of coalition formation principles, tasks decomposition and distribution,

winnings sharing, and implementation of proposed techniques and models. This work focuses on ensuring the

interaction of coalition members through distributed ledger technology and smart contracts using Hyperledger

Fabric platform, as well as modeling and visualizing the interaction of intelligent robots using open software

Gazebo and Robotic Operation System. The ontology of context used to adjust robot actions is presented. It

combines environmental characteristics with robots and tasks descriptions to provide full situation context.

The paper presents a modelling approach architecture with an example of modelling and visualization based

on obstacle overcoming scenario.

1 INTRODUCTION

Оne of the main directions in the field of collective

work of robots is the study of interaction models of

intelligent agents (Bayram & Bozma, 2015; Vig &

Adams, 2005). Interaction of agents is most often

required when solving a problem that cannot be

solved by the efforts of one agent due to the lack of

its capabilities. In this case, the task is divided into

several independent subtasks. Each of them is then

assigned to a separate agent (Bayram & Bozma, 2015;

Cui, Guo, & Gao, 2013). The importance and

relevance of these studies is due to the development

of robotic systems in which each individual robot can

be considered as an independent intelligent agent. A

robot can make decisions to achieve its own goals and

a common goal. This interaction model is also called

a coalition. A wide range of tasks in the development

of coalition interaction models lies in the field of

models that provide the most optimal choice of

participants in a collective solution to the problem

and ensure the interaction of coalition members

(Dukeman & Adams, 2017; Klusch & Gerber, 2002).

The formation of a coalition and the joint solution

of tasks by robots can be demonstrated in most detail

within the following subject areas: precision

https://orcid.org/0000-0003-0619-8620

agriculture, remote planet exploration, emergency

medicine. In all these areas, many high specialized

robots have been developed. They effectively solve a

limited set of tasks within the framework of the

conditions for which they were developed. For

example, to solve the problem of precision

agriculture, there are robots that can conduct soil

quality exploration, automated seeders, cultivators,

tractors, etc.

However, to solve a complex problem that goes

beyond the conditions of each robot, it is required to

provide platform for their joint work. To ensure the

full cycle of growing crops, it is necessary to

successively solve the problems of field exploration,

selection of crops grown, sowing, watering and

harvesting. To do this, it is required to form a

coalition of robots in such a way as to solve a complex

common problem with the maximum overall

effectiveness of the coalition. Efficiency calculating

for different subject areas depends on the gain that

can be obtained when solving the problem: the

maximum harvest for precision agriculture, the

number of people rescued for disaster medicine, and

the total amount and time spent on obtaining,

processing and transmitting information about the

state of the monitored objects of interest for remote

sensing tasks.

278

Smirnov, A. and Teslya, N.

Modelling and Visualization of Robot Coalition Interaction through Smart Space and Blockchain.

DOI: 10.5220/0009413102780286

In Proceedings of the 22nd International Conference on Enterprise Information Systems (ICEIS 2020) - Volume 2, pages 278-286

ISBN: 978-989-758-423-7

The solution of the coalition formation problem

with the mathematical theory of fuzzy cooperative

games was presented in a previous work by the

authors (Smirnov, Sheremetov, & Teslya, 2019). This

work focuses on the secure interaction of coalition

members through distributed ledger technology and

smart contracts using Hyperledger Fabric platform, as

well as modeling and visualizing the interaction of

intelligent robots with open software Gazebo and

Robotic Operation System.

The paper is structured as follows. Section 2

discusses the existing forms of coalitions and

methods for modeling and visualizing their

interaction. Section 3 presents a coalition interaction

model based on the concept of smart spaces,

including a description of the ontology-based context

model used to form the coalition. Section 4 presents

the architecture for the coalition members interaction

through a distributed ledger. Section 5 shows the

architecture and an example of visualizing the

interaction of coalition members using robots’

models and simulating the physical world in the

Gazebo environment. Section 6 concludes the article

and provides directions of future work.

2 RELATED WORK

Currently, the main research directions of solving the

problem of coalition work of intelligent robots are the

following: organization of a coalition, information

exchange between coalition members, distribution of

tasks and resources between coalition members.

Two major areas can be distinguished in the

process of a coalition organizing: centralized and

decentralized. The centralized organization of the

coalition is characterized by the presence of a control

center at which decisions are made on the

composition of the coalition, the distribution of tasks,

and forming a plan for solving the problem. In this

case, the control center can be represented either as a

separate high-powered computing device that

performs only the functions of the center, and a robot

that performs tasks along with the rest (Smirnov,

Kashevnik, & Ponomarev, 2015). In this case, the

structure of the coalition can be multilevel

hierarchical, in which robots at each underlying level

obey only one center of their superior level (Guerrero,

Oliver, & Valero, 2017). The decentralized

organization of the coalition usually implies the

absence of a decision center, often focusing on bio-

inspired methods of organizing collaboration, such as

swarms and flocks (Koes, Nourbakhsh, & Sycara,

2005; Yu & Cai, 2009). At the same time, robots

equal in hierarchy and all of them use same

algorithms when making a decision.

The information exchange between coalition

members is an important component of the joint

problem solving by coalition, since it requires

notification of coalition members about the current

state of the problem solution for organizing

coordinated actions, or monitoring the

implementation of the plan (Verma, Desai, Preece, &

Taylor, 2017). The information exchange can be

organized through a common centralized repository

of information on a separate device or by the

distribution of information between coalition

members (Shabanov & Ivanov, 2019). It also

considers the combination of two approaches with the

formation of so-called smart spaces - a common

repository of information that provides links to

resources that are coalition members, which makes

information distributed among all participants

(Ferrer, 2019). There are also solutions based on

peer2peer networks and a distributed ledger that

provide quick distribution of information between all

participants, while duplicating all the information on

the device of each participant (Liang & Xiao, 2010;

Qian & Cheng, 2018). The distributed ledger

technology can be viewed as an example of P2P

networks. In addition to benefits of P2P network, the

distributed ledger solutions also provide immutability

of information, which can be useful when organizing

a coalition with the requirement to ensure trust

between the participants without a single certification

center (Liang & Xiao, 2010).

The type of coalition organization also influences

how tasks and resources are distributed among

coalition members. Centralized hierarchical

coalitions usually organize work through centralized

planning when receiving a task. At higher nodes, a

work plan is built taking into account the capabilities

of lower nodes and coalition resources, in which the

performers and the procedure for solving problems

are fixed (Smirnov, Kashevnik, Teslya, Mikhailov, &

Shabaev, 2015). This ensures that the stages of the

plan and the entire plan are completed by a certain

date with an accurate forecast of the expenditure of

resources and the payment of remuneration, if this is

provided for by the conditions of the task. However,

this solution is not flexible, because when an

emergency occurs, it leads to a deviation from the

plan, with the need for its correction or complete

reorganization. Decentralized coalitions are based on

the adaptation of participants to current conditions,

with the absence of a single plan for solving the

problem (Hartanto & Eich, 2014; Tosello, Fan,

Castro, & Pagello, 2017). This provides the flexibility

Modelling and Visualization of Robot Coalition Interaction through Smart Space and Blockchain

279

to solve the problem under frequent changes in the

composition of the coalition or available resources,

but limits the ability to predict the time of solving the

problem (Ivanov, 2019).

Simulation and visual modeling of the interaction

of robots is also very important task. Both of them are

greatly simplifies hypothesis testing by reducing

development costs, and allow to visually present the

results of the proposed approaches. Some researchers

develop their own visualizations, displaying robots

with conventional signs, since the tasks under

consideration do not require detailed design of the

robot (Koes et al., 2005). For detailed visualization,

the Gazebo visualization package is most often used

in combination with the control code of the robot

operating system (Robot operation system, ROS) for

controlling a virtual robot (Barbosa, Duberg, Jensfelt,

& Tumova, 2019; Suárez-Figueroa, 2012; Xue, Tang,

Su, & Li, 2019).

In summary it can be noted that nowadays the

main attention is paid to the creation of decentralized

coalitions of robots to solve complex common

problems. The distribution of tasks and resources

between them should be carried out based on

decentralized planning to adapt to the changing

context of the task, while robots take part in solving

the problem considering their functionality and

available resources, such as battery power or device

lifetime. The interaction and exchange of information

between robots should be carried out through a

common repository. Taking into account the recent

studies two approaches towards decentralization can

be viewed: P2P model, and the use of distributed

ledgers. If it is not possible to test interaction models

on real robots, visualization environments should be

used to model robots and their actions. The most

commonly used bundle is the Gazebo simulation

environment, together with the robotics operating

system (ROS), which implements the robots

functionality.

3 INTERACTION OF

COALITION PARTICIPANTS

Robots interact through the cyberphysical framework

described previously in work (Smirnov et al., 2019).

The framework is based on the smart cyberphysical

space (based on the “blackboard”) and blockchain. It

provides the ability to organize basic interaction of

robots in the physical and cyber (virtual) spaces. The

interaction includes solo and joint manipulations with

physical objects, information exchange about the

current state of robots and objects for planning further

joint actions during the coalition formation.

Coalition members can be robots of different

manufacturers with different equipment,

environmental sensors, and software modules. For

their interaction, it is important to ensure semantic

interoperability. This allows to specify a description

of the properties of the coalition members and context

in a format that is understandable to everyone, from

machines to the system’s operators, as well as

automate the search for coalition members in

accordance with the requirements for the task. An

ontology is used to ensure semantic interoperability.

The model of the context, the coalition participant and

tasks is described using the ontological modeling

apparatus, which allows context-driven dynamic

formation of the coalition and the distribution

(redistribution) of tasks, roles and system resources

taking into account not only competencies, but also

the current situation in the coalition.

To build an ontological model of the context, the

main scenarios of robots interaction, robot designs,

and typical tasks that each type of robot solves for

disaster medicine, precision farming, and remote

sensing of the Earth had been analyzed. As a result,

the basic concepts that should be present in the

ontology were identified, among which three groups

stand out: i) concepts for the current situation; ii)

concepts for the design and functions of the robot; iii)

concepts for the requirements and conditions for

solving specific problems posed to a given group of

robots.

The ontology graph for context is presented on

Figure 1. For the current situation, the concepts

describe the physical parameters of the environment

(for example, temperature, wind speed and direction,

humidity, current time, atmospheric pressure for

precision farming and disaster medicine, for remote

sensing - the tactical and technical characteristics of

onboard target and supporting equipment, potential

areas of interaction with objects observation and

ground-based points for receiving Earth remote

sensing data, light level, solar activity, radiation level,

quality indicators of function observation of

spacecraft, etc.), the position of all the coalition

robots, the position and properties of objects in the

field of action of the coalition.

Ontology concepts for describing the design and

functions of the robot provide characteristics of the

hardware and software components of the robot

(number, types, measuring ranges and current sensor

readings, number, types, current position of motors,

state and current battery charge, fuel level (if there is

fuel tank) computing power (CPU frequency, the

ICEIS 2020 - 22nd International Conference on Enterprise Information Systems

280

Figure 1: Context ontology.

amount of RAM, data storage, connection to data

networks, information about the characteristics of the

selected GNSS)) and many components (installed

libraries, software modules, platforms, control code

for hardware, action templates), as well as a

description of the functions and possible technologies

that the robot can perform using its hardware and

software. The concepts of the developed ontology are

synchronized with the ontology defined by the

standard 1872-2015 - IEEE Standard Ontologies for

Robotics and Automation (IEEE Robotics and

Automation Society, 2015) to ensure interoperability

while expanding the range of subject areas and the

composition of coalition members.

Ontology concepts related to the task description

and task requirements determine the types of tasks,

the resources needed to solve them, the requirements

for the functional equipment of robots, as well as the

possible structures for dividing tasks solved by

coalition of robots into sub-tasks.

4 COALITION MEMBERS

INTERACTION THROUGH

BLOCKCHAIN

Some platforms that implement blockchain

technology can be extended by the use of smart

contracts to provide new capabilities of ledger

processing. For the purpose of coalition participant

interaction, a smart contract within blockchain

technology is viewed as a decentralized application

available to all coalition participants.

In this work the Hyperledger Fabric platform has

been chosen for blockchain network and smart

contacts implementation. The choice is justified by

the specifics of architecture, which makes it easy to

adapt the coalition structure into the platform

structure (see Fig. 2). The main elements of the

architecture are nodes of three levels: “Client”,

“Peer”, “Orderer”. Client level corresponds to robots

whose main task is to send data from sensors, or to

perform operations and report on their performance.

In case of precision farming, such robots can be tools

of combine harvester, scouts, and transport robots. On

the higher lever, there are devices that collect

information and execute the smart contracts - “Peer”.

The example of peer is a control block of combine

harvester. Their main task is to collect information

from the lower level, process it using smart contracts

and transfer it to the upper level, in which information

will be disseminated and stored. The highest level is

“Orderer”. Its task is to store information in the

appropriate chain of blocks, to ensure the

coordination and distribution of the new block

between other Orderers and corresponding Peers.

All changes of coalition state are reflected in the

smart space using adding/deleting relevant

information according to the in the ontology. During

the interaction between coalition participants through

the IoT platform with the blockchain support, all

smart contacts can be called either directly through

the transaction initiation interface in the blockchain

Modelling and Visualization of Robot Coalition Interaction through Smart Space and Blockchain

281

or using other smart contacts methods. To simplify

the use of custom protocols in blockchain, the method

of any contract placed in the contract chains should

be called using the basic smart contract.

For this purpose, a basic smart contract was

developed, which provides the following functions of

interaction between coalition participants:

1. Coalition schedule upload using XML format.

It provides receiving and parsing the plan; generating

entries in the blockchain from the extracted items of

robots, tasks associated with robots, the order of the

task execution, and the timing of each task; sending

notification to the blackboard through blockchain

knowledge processor to start the plan execution.

2. Start the task execution. Accept a message from

the robot about the start of the task execution, store

the moment of the real start of execution in the

blockchain, verify with the planned one, and store the

fact in case of a strong deviation.

3. Completion of the task. Same as with the start

the contract provides function to get notification,

check correctness of execution and store this fact to

the ledger. In case of a strong deviation from the plan,

this fact is stored to the blockchain and notification

about the failure of the plan is sent to other coalition

members through the smart space.

The basic smart contract is available through the

REST API so each robot can access the schedule

without direct connection to the blockchain just using

HTTP protocol. In addition, a simple web page is

available for a human operator to check the coalition

state and follow the process of joint task solving.

5 VISUALIZATION

ARCHITECTURE AND

COALITION INTERACTION

EXAMPLE

For the experiments, a scenario of coalitional

interaction of robots was implemented to jointly

overcome obstacles in a Gazebo modeling

environment (Agüero et al., 2015; Koenig & Howard,

2004) with ROS Melodic framework (Stanford

Artificial Intelligence Laboratory et.al., 2018) (see

Figure 3). The use of the above software is currently

widespread in the field of robotic modeling and

includes a diverse library of elements and off-the-

shelf devices, which can significantly reduce model

development time. Due to the possibility of

connecting third-party models, it is also possible to

quickly create a three-dimensional physical world

that simulates the real world with automatic object

collision calculation, environmental physical

parameters. This feature allows to faithfully display

the physical world and take into account a large

number of parameters during modeling.

Using the ROS operating system also has several

advantages. The first is that this operating system

supports the basic functions of real-life robots related

Orderer

HyperLedger Fabric

blockchain with Smart Space

Coalition 1

Main Robot

Figure 2: Coalition members’ interaction through HyperLedger Fabric with Smart Space support.

Correspondence of coalition members with Hyperledger concepts

Interaction through blockchain and Smart Space

Interaction through smart contracts call

Gazebo

Robot Client 1

ROS

KP client

Channels

Robot Peer 1

ROS

KP Pee

Channels

Coalition n

Main

Robot Client n

ROS

KP client

Channels

Robot Peer n

ROS

KP Pee

Channels

draw

Orderer

Sensors data,

tasks, schedule

Current state

Sensors data,

tasks, schedule

ICEIS 2020 - 22nd International Conference on Enterprise Information Systems

282

to controlling servo drives, receiving data from

sensors and exchanging data with the external

environment through an accessible communication

channel. Interaction between robots is carried out

through specially organized channels in the operating

system, to which a program can be subscribed and

receive sensor readings from them or send control

commands. In the presented architecture, the robot

knowledge processor (KP) subscribes to channels and

transmits messages from the robot to the smart space

and blockchain using the ontology, as well as receives

back the task that needs to be completed, and the

schedule and controls the actions of the robot.

The second important advantage is the ability to

integrate ROS both in existing robots and in robot

models created in the Gazebo simulation

environment. This allows to develop a model of

robots interaction in the virtual world and then, with

minimal changes, transfer it to physical robots, and

due to high-quality simulation of the physical world

conditions, many features of the physical

environment will be taken into account during virtual

implementation and estimation.

The following robot models were developed to

implement the scenario (see Figure 4): all-terrain

robot, quadcopter, and scout.

All-terrain robot has the following characteristics:

six wheels, three independent parts (front, center and

back). The front and back parts have lift mechanisms

that allows to rise or down parts of robots to overcome

obstacles. On the front and back parts of the robot

there are laser distance sensors to measure the

distance to objects, directed in parallel to the surface

on which the robot moves. The robot is equipped with

a coordinate sensor (GPS/GLONASS in real robot)

for tracking location.

Quadcopter robot has the functionality of a

standard quadcopter for moving in three-dimensional

space. The robot is equipped with a laser distance

sensor to scan for obstacles and is directed downward

perpendicular to the plane of rotation of the rotors.

Same as all-terrain robot the quadcopter is equipped

with a coordinate sensor for tracking location. The

quadcopter model, which implements the quadcopter

robot functionality, was implemented based of the

hector_quadrotor free model package (Meyer,

Sendobry, Kohlbrecher, Klingauf, & Von Stryk,

2012, 2018).

Scout robot is a four-wheel mobile robot with the

function of moving on a two-dimensional surface

Robotic Operation System

Gazebo Simulator

gazebo_ros_pkgs (interface for

connecting ROS and Gazebo

simulator)

Quadcopter

All-terrain robot

.launch files

Quadcopter model (SDF and

URDF)

All-terrain robot model (SDF

and URDF)

Environment and obstacle

model (SDF)

Control modules /

User software

goToLocation.py

publishTask.py

scanObject.py

Message channels

Location

Quadcopter

All-terrain robot

Movement

Quadcopter

All-terrain robot

Task scheduler

Sensors

uadco

All-terrain robo

Robots and world

description

Visualization interface

Scout

Figure 3: Software architecture for robot interaction visualization.

Modelling and Visualization of Robot Coalition Interaction through Smart Space and Blockchain

283

without huge obstacles. The robot is equipped with

laser distance sensors located around the perimeter of

the body to scan the space around the robot. The robot

is also equipped with a coordinate sensor for tracking

location. Scout robot model was implemented based

on the husky_gazebo package (Gariepy, Mukherjee,

Bovbel, & Ash, 2019).

The transmission of control signals to the robot

and the receipt of data from the sensors is carried out

through subscription to the channels in ROS

associated with the equipment of the robots. For

example, for the all-terrain robot, channels have been

created through which data from the distance sensor

and coordinates can be received and independent

control actions to each of the robot servos can be sent.

For each of the servos, it is also possible to obtain its

status by accessing the corresponding channel.

6 CONCLUSIONS

The paper provides modelling of the scenario of joint

problem solving by robots’ coalition proposed early

by the authors. The modelling is based on the open

source packages Gazebo and ROS that are de-facto

standards of virtual environments modelling in

visualization in the field of robotics. The interaction

between robots is implemented based on the smart

space combined with the blockchain platform.

To describe the environment the ontological

model of context has been developed. The ontological

model provides the concretization of the parameters

that can be used to describe the context of the joint

problem solving in various fields, including precision

agriculture, disaster medicine, and remote sensing of

the Earth. The benefit of ontological context model is

in easy adaptation to other areas by supplementing it

with concepts that describe the current situation,

specific to these areas, while maintaining the context

in terms of describing the structure and functions of

robots and tasks.

The visualization shows simple scenario of three

robot cooperation for obstacle overcoming. Selection

of tools for visualization and communication between

robots shows how they can be combined for

modelling scenarios of robot cooperation. This result

can be further scaled to implement more complex

scenario of precision agriculture that requires

implementation of physical world, robot models, and

providing secure interaction between them. The

interaction will include secured interoperation

through blockchain and cyber-physical space,

coalition formation, and rescheduling of complex

problem solving.

The future work will be focused in precision

agriculture scenario implementation and performance

estimation to prove that the proposed solution is

appropriate for this kind of tasks.

ACKNOWLEDGEMENTS

This research was funded by Russian State Research

No. 0073-2019-0005 in the sections 2 and 3 of cyber-

physical system organization, and by RFBR Grant

No. 17-29-07073 in the sections 4 and 5 of robot

interoperation through blockchain.

REFERENCES

Agüero, C. E., Koenig, N., Chen, I., Boyer, H., Peters, S.,

Hsu, J., … Pratt, G. (2015). Inside the Virtual Robotics

Challenge: Simulating Real-Time Robotic Disaster

Response. IEEE Transactions on Automation Science

and Engineering, 12(2), 494–506. https://doi.org/

10.1109/TASE.2014.2368997

Quadcopter

Scou

All-terrain robot

Obstacle model

Figure 4: Visualization of robots interaction with Gazebo and ROS (left side– initial scenario stage, right side – final stage).

ICEIS 2020 - 22nd International Conference on Enterprise Information Systems

284

Barbosa, F. S., Duberg, D., Jensfelt, P., & Tumova, J.

(2019). Guiding Autonomous Exploration With Signal

Temporal Logic. IEEE Robotics and Automation

Letters, 4(4), 3332–3339. https://doi.org/

10.1109/lra.2019.2926669

Bayram, H., & Bozma, H. I. (2015). Coalition formation

games for dynamic multirobot tasks. International

Journal of Robotics Research, 35(5), 514–527.

https://doi.org/10.1177/0278364915595707

Cui, R., Guo, J., & Gao, B. (2013). Game theory-based

negotiation for multiple robots task allocation.

Robotica, 31(6), 923–934. https://doi.org/10.1017/

S0263574713000192

Dukeman, A., & Adams, J. A. (2017). Hybrid mission

planning with coalition formation. Autonomous Agents

and Multi-Agent Systems, 31(6), 1424–1466.

https://doi.org/10.1007/s10458-017-9367-7

Ferrer, E. C. (2019). The blockchain: A new framework for

robotic swarm systems. Advances in Intelligent Systems

and Computing, 881, 1037–1058. https://doi.org/

10.1007/978-3-030-02683-7_77

Gariepy, R., Mukherjee, P., Bovbel, P., & Ash, D. (2019).

GitHub - husky/husky: Common packages for the

Clearpath Husky. Retrieved January 6, 2020, from

https://github.com/husky/husky

Guerrero, J., Oliver, G., & Valero, O. (2017). Multi-Robot

Coalitions Formation with Deadlines: Complexity

Analysis and Solutions. PLOS ONE, 12(1), 1–26.

https://doi.org/10.1371/journal.pone.0170659

Hartanto, R., & Eich, M. (2014). Reliable, cloud-based

communication for multi-robot systems. 2014 IEEE

International Conference on Technologies for Practical

Robot Applications (TePRA), 1–8. https://doi.org/

10.1109/TePRA.2014.6869142

IEEE Robotics and Automation Society. (2015). IEEE

Standard Ontologies for Robotics and Automation.

https://doi.org/10.1109/IEEESTD.2015.7084073

Ivanov, D. (2019). Decentralized planning of intelligent

mobile robot’s behavior in a group with limited

communications. In Advances in Intelligent Systems

and Computing (Vol. 875). https://doi.org/

10.1007/978-3-030-01821-4_44

Klusch, M., & Gerber, A. (2002). Dynamic coalition

formation among rational agents. IEEE Intelligent

Systems, 17(3), 42–47. https://doi.org/10.1109/

MIS.2002.1005630

Koenig, N., & Howard, A. (2004). Design and use

paradigms for Gazebo, an open-source multi-robot

simulator. 2004 IEEE/RSJ International Conference on

Intelligent Robots and Systems (IROS), 3, 2149–2154.

https://doi.org/10.1109/iros.2004.1389727

Koes, M., Nourbakhsh, I., & Sycara, K. (2005).

Heterogeneous multirobot coordination with spatial and

temporal constraints. AAAI Workshop - Technical

Report,

WS-05-06, 9–16.

Liang, X., & Xiao, Y. (2010). Studying bio-Inspired

coalition formation of robots for detecting intrusions

using game theory. IEEE Transactions on Systems,

Man, and Cybernetics, Part B: Cybernetics, 40(3),

683–693. https://doi.org/10.1109/

TSMCB.2009.2034976

Meyer, J., Sendobry, A., Kohlbrecher, S., Klingauf, U., &

Von Stryk, O. (2012). Comprehensive simulation of

quadrotor UAVs using ROS and Gazebo. Lecture Notes

in Computer Science (Including Subseries Lecture

Notes in Artificial Intelligence and Lecture Notes in

Bioinformatics), 7628 LNAI, 400–411. https://doi.org/

10.1007/978-3-642-34327-8_36

Meyer, J., Sendobry, A., Kohlbrecher, S., Klingauf, U., &

Von Stryk, O. (2018). GitHub - tu-darmstadt-ros-

pkg/hector_quadrotor: hector_quadrotor contains

packages related to modeling, control and simulation of

quadrotor UAV systems. Retrieved January 5, 2020,

from https://github.com/tu-darmstadt-ros-pkg/

hector_quadrotor

Qian, B., & Cheng, H. H. (2018). Bio-Inspired Coalition

Formation Algorithms for Multirobot Systems. Journal

of Computing and Information Science in Engineering,

18(2), 1–8. https://doi.org/10.1115/1.4039638

Shabanov, V., & Ivanov, D. (2019). Organization of

information exchange in coalitions of intelligent mobile

robots. 2019 International Conference on Industrial

Engineering, Applications and Manufacturing,

ICIEAM 2019, 1–5. https://doi.org/10.1109/

ICIEAM.2019.8743043

Smirnov, A., Kashevnik, A., & Ponomarev, A. (2015).

Multi-level self-organization in cyber-physical-social

systems: Smart home cleaning scenario. Procedia

CIRP, 30, 329–334. https://doi.org/10.1016/

j.procir.2015.02.089

Smirnov, A., Kashevnik, A., Teslya, N., Mikhailov, S., &

Shabaev, A. (2015). Smart-M3-based robots self-

organization in pick-and-place system. 2015 17th

Conference of Open Innovations Association (FRUCT),

2015-June(June), 210–215. https://doi.org/10.1109/

FRUCT.2015.7117994

Smirnov, A., Sheremetov, L., & Teslya, N. (2019). Fuzzy

cooperative games usage in smart contracts for dynamic

robot coalition formation: Approach and use case

description. ICEIS 2019 - Proceedings of the 21st

International Conference on Enterprise Information

Systems, 1, 349–358. https://doi.org/10.5220/

0007763003610370

Stanford Artificial Intelligence Laboratory et.al. (2018).

ROS.org | Powering the world’s robots. Retrieved

January 5, 2020, from https://www.ros.org/

Suárez-Figueroa, M. C. (2012). Ontology engineering in a

networked world. Springer.

Tosello, E., Fan, Z., Castro, A. G., & Pagello, E. (2017).

Cloud-Based Task Planning for Smart Robots. In

Intelligent Autonomous Systems 14, Advances in

Intelligent Systems and Computing (Vol. 531, pp. 285–

300). https://doi.org/10.1007/978-3-319-48036-7_21

Verma, D., Desai, N., Preece, A., & Taylor, I. (2017). A

block chain based architecture for asset management in

coalition operations. In T. Pham & M. A. Kolodny

(Eds.), Proc. SPIE 10190, Ground/Air Multisensor

Interoperability, Integration, and Networking for

Modelling and Visualization of Robot Coalition Interaction through Smart Space and Blockchain

285

Persistent ISR VIII (p. 101900Y). https://doi.org/

10.1117/12.2264911

Vig, L., & Adams, J. (2005). Issues in multi-robot coalition

formation. Multi-Robot Systems. From Swarms to

Intelligent Automata Volume III, III, 15–26.

https://doi.org/10.1007/1-4020-3389-3_2

Xue, F., Tang, H., Su, Q., & Li, T. (2019). Task Allocation

of Intelligent Warehouse Picking System based on

Multi-robot Coalition. KSII Transactions on Internet

and Information Systems, 13(7), 3566–3582.

https://doi.org/10.3837/tiis.2019.07.013

Yu, L., & Cai, Z. (2009). Robot exploration mission

planning based on heterogeneous interactive cultural

hybrid algorithm. 5th International Conference on

Natural Computation, ICNC 2009, 5, 583–587.

https://doi.org/10.1109/ICNC.2009.15

ICEIS 2020 - 22nd International Conference on Enterprise Information Systems

286