ArticlePDF Available

Collaborative Robotics Research: Subiko Project

Authors:
  • NextTechnologies Kft
  • NextTechnologies Ltd

Abstract and Figures

Aim of the research is to present the possibilities of applying cooperative robots during the process of automotive metalworking. The study is focusing at the Hungarian medium enterprise sector. Artificial Intelligence and special cobot safety systems - modified by human behavior - are used to demonstrate how production techniques are used at a Hungarian medium enterprise to optimize their processes. The main problem is with flexibility in the automotive metalworking manufacturing industry, such as production line switchover and the processing period. The product price is therefore determined by the competition, and the only way to increase profit is to reduce production and distribution costs. This means that managing and operating the organization and manufacturing in an efficient manner is necessary. One of the success factors is the flexibility of manufacturing by robotization. The proposal solution by this study is a low-load universal cobot system with innovative security solutions for improve the flexibility of manufacturing in an automotive metalworking manufacturing company. This instance is based on a real case study problem in an automotive metalworking manufacturing company.
Content may be subject to copyright.
ScienceDirect
Available online at www.sciencedirect.com
Procedia Manufacturing 46 (2020) 467–474
2351-9789 © 2020 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer-review under responsibility of the scientific committee of the 13th International Conference Interdisciplinarity in Engineering.
10.1016/j.promfg.2020.03.068
10.1016/j.promfg.2020.03.068 2351-9789
© 2020 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer-review under responsibility of the scientic committee of the 13th International Conference Interdisciplinarity in Engineering.
Available online at www.sciencedirect.com
ScienceDirect
Procedia Manufacturing 00 (2019) 000–000
w
ww.elsevier.com/locate/procedia
2351-9789 © 2019 The Authors. Published by Elsevier B.V.. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer-review under responsibility of the scientific committee of the 13th International Conference Interdisciplinarity in Engineering
13th International Conference Interdisciplinarity in Engineering (INTER-ENG 2019)
Collaborative Robotics Research: Subiko
Project
Daniel Tokody
a,
*, Laszlo Ady
b
, Luca F. Hudasi
c
, Péter János Varga
b
, Péter Hell
a
a
Doctoral School on Safety and Security Sciences, Óbuda University, Budapest, Népszínház utca 8., 1081, Hungary
b
Kálmán Kandó Faculty of Electrical Engineering, Óbuda University, Tavaszmező utca 15-17., 1084, Hungary
c
NextTechnologies Ltd., Maglód, Sugár út 44., 2234, Hungary
Abstract
Aim of the research is to present the possibilities of applying cooperative robots during the process of automotive metalworking.
The study is focusing at the Hungarian medium enterprise sector. Artificial Intelligence and special cobot safety systems -
modified by human behavior - are used to demonstrate how production techniques are used at a Hungarian medium enterprise to
optimize their processes. The main problem is with flexibility in the automotive metalworking manufacturing industry, such as
production line switchover and the processing period. The product price is therefore determined by the competition, and the only
way to increase profit is to reduce production and distribution costs. This means that managing and operating the organization
and manufacturing in an efficient manner is necessary. One of the success factors is the flexibility of manufacturing by
robotization. The proposal solution by this study is a low-load universal cobot system with innovative security solutions for
improve the flexibility of manufacturing in an automotive metalworking manufacturing company. This instance is based on a real
case study problem in an automotive metalworking manufacturing company.
© 2019 The Authors. Published by Elsevier B.V.. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer-review under responsibility of the scientific committee of the 13th International Conference Interdisciplinarity in
Engineering
Keywords: Cooperative robots; Human-robot interaction; Safety-critical systems; R&D; Automotive metalworking.
* Corresponding author. Tel.: +36-309-507-193.
E-mail address: daniel_tokody@ieee.org
Available online at www.sciencedirect.com
ScienceDirect
Procedia Manufacturing 00 (2019) 000–000
w
ww.elsevier.com/locate/procedia
2351-9789 © 2019 The Authors. Published by Elsevier B.V.. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer-review under responsibility of the scientific committee of the 13th International Conference Interdisciplinarity in Engineering
13th International Conference Interdisciplinarity in Engineering (INTER-ENG 2019)
Collaborative Robotics Research: Subiko
Project
Daniel Tokody
a,
*, Laszlo Ady
b
, Luca F. Hudasi
c
, Péter János Varga
b
, Péter Hell
a
a
Doctoral School on Safety and Security Sciences, Óbuda University, Budapest, Népszínház utca 8., 1081, Hungary
b
Kálmán Kandó Faculty of Electrical Engineering, Óbuda University, Tavaszmező utca 15-17., 1084, Hungary
c
NextTechnologies Ltd., Maglód, Sugár út 44., 2234, Hungary
Abstract
Aim of the research is to present the possibilities of applying cooperative robots during the process of automotive metalworking.
The study is focusing at the Hungarian medium enterprise sector. Artificial Intelligence and special cobot safety systems -
modified by human behavior - are used to demonstrate how production techniques are used at a Hungarian medium enterprise to
optimize their processes. The main problem is with flexibility in the automotive metalworking manufacturing industry, such as
production line switchover and the processing period. The product price is therefore determined by the competition, and the only
way to increase profit is to reduce production and distribution costs. This means that managing and operating the organization
and manufacturing in an efficient manner is necessary. One of the success factors is the flexibility of manufacturing by
robotization. The proposal solution by this study is a low-load universal cobot system with innovative security solutions for
improve the flexibility of manufacturing in an automotive metalworking manufacturing company. This instance is based on a real
case study problem in an automotive metalworking manufacturing company.
© 2019 The Authors. Published by Elsevier B.V.. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer-review under responsibility of the scientific committee of the 13th International Conference Interdisciplinarity in
Engineering
Keywords: Cooperative robots; Human-robot interaction; Safety-critical systems; R&D; Automotive metalworking.
* Corresponding author. Tel.: +36-309-507-193.
E-mail address: daniel_tokody@ieee.org
468 Daniel Tokody et al. / Procedia Manufacturing 46 (2020) 467–474
2 Daniel Tokody et al. / Procedia Manufacturing 00 (2019) 000–000
1. Introduction
The development and application of collaborative robots is progressing more and more dynamically. This is due
to the fact that such robots can be flexibly integrated into a workflow where robots and humans are required to work
in a common work space, with well-defined protection zones. In these workflows, the user - the human - controls the
pre-recorded work of the cooperative robot. In many cases, the robot performs the stressful and monotonous part of
the task in the collaboration. This technology creates a new kind of risk and requires updating and developing new
systems of requirements, laws and standards. [1][2][3][4][5]
1.1. Application of Collaborative Robots
In a manufacturing process (e.g. powertrain assembly [6]) combined with a general robot, the robots are located
in a space enclosed by protective fences and sensors and are physically separated from humans. The new goal is to
secure a common working space for cobot and man.
1.2. Perception of Robot Danger
In any case, it is up to the employer to decide whether the collaborative robot is a dangerous work tool, there is
currently no legal requirement for this, but guidance can be provided by the Hungarian Labour Protection Act. [7] It
is necessary to consider the maximum force [8] (motion control: desired forces/torques [9]) that a cobot can exert on
a human in the event of a collision.
1.3. Legislative background to the use of robots
Collaborative robots have a very short history (this research topic also has regulatory dilemmas [10]), so there are
few international standards adapted to the Hungarian language. Therefore, international standards are applicable
when integrating robots into the workflow. The standards currently known and used are as follows (see Table 1):
1.4. Terms and definitions
System Engineering: “At NASA, “systems engineering” is defined as a methodical, multi-disciplinary approach
for the design, realization, technical management, operations, and retirement of a system. A “system” is the
combination of elements that function together to produce the capability required to meet a need. The elements
include all hardware, software, equipment, facilities, personnel, processes, and procedures needed for this purpose;
that is, all things required to produce system-level results.” [11]
Collaborative workspace: “Collaborative workspace space within the operating space where the robot system
(including the workpiece) and a human can perform tasks concurrently during production operation”[12]
Collaborative mode: The state in which the robot is working with the robot operator in the collaborative
workspace. [12]
Safety distance: the minimum allowable distance between the robot and the operator. [12]
Fail operational: systems are able to operate even if their control system fails. [13]
Fail-safe: systems mean that in the event of a system failure, the system becomes safe preventing the occurrence
of more serious problems. [13]
2. Collaboration Levels for Human-Robot Cooperation
There are multiple levels of interaction between industrial robots and their operator. In manual handover mode,
the robot communicates directly with the operator in the handover window, where the robot has reduced speed and a
working space with inhibited sensors. Through the window, the operator can safely change workpieces during
automatic operation. A more advanced form of this is the contact window when the robot stops at the contact
window and the operator can manually move it out of the workspace delimited by the window. This is called Hold-
Daniel Tokody et al. / Procedia Manufacturing 46 (2020) 467–474 469
Daniel Tokody et al. / Procedia Manufacturing 00 (2019) 000–000 3
To-Run control mode. In another mode of application, the robot first decreases its speed in proportion to the distance
between the operator and the robot, and then stops when a person enters a restricted zone of his work area. The robot
can be restarted if the forbidden zone is empty. In the case of the most advanced cooperative setup, the operator can
move the robot with manual control, in a specific common work space, on a specific path. Previously, robot-based
technologies (such as robotic cells) had been used to separate human and machine work from one another, while
recent innovations aim to break these boundaries and create cooperative and collaborative operations.
Table 1. The important standards for Research and Development of Cobots (own edit)
ISO/TS 15066:2016
Robots and robotic devices
Collaborative robots
ISO 10218-1:2011 Safety requirements for industrial robots - Part 1: Robots
ISO 10218-2:2011 Safety requirements for industrial robots - Part 2: Robot systems and integration
IEC 61508-1:2010
Functional safety of
electrical/electronic/ programmabl
e electronic safety-related systems
Part 1: General requi rements
IEC 61508-2:2010 Part 2: Requirements for electrical/electronic/programmable electronic safety-related systems
IEC 61508-3:2010 Part 3: Software requirement s
IEC 61508-4:2010 Part 4: Definitions and abbreviations
IEC 61508-5:2010 Part 5: Examples of met hods for the determination of safety integrity levels
IEC 61508-6:2010 Part 6: Guidelines on the application of IEC 61508-2 and IEC 61508-3
IEC 61508-7:2010 Part 7: Overview of techniques and measures
ISO 12100:2010
Safety of machinery
General principles for design - Risk assessment and risk reduction
ISO 13850:2015 Emergency stop function - Principles for design
ISO 13855:2010 Positioning of safeguards with respect to the approach speeds of parts of the human body
ISO 13482:2014
Robots and robotic devices
Safety requirements for personal care robots
ISO 8373:2012 Vocabulary
IEC 62061:2005
Safety of machinery
Functional safety of safety-related electrical, electronic and programmable electronic control
systems
ISO 13849-1:2015 Safety-related parts of control systems - Part 1: General principles for design
ISO 13849-2:2012 Safety-related parts of control systems - Part 2: Validation
IEC 62443-4-1:2018 Security for industrial automation
and control systems Part 4-1: Secure produc t development lifecycle requirements
IEC 62443-2-1:2010
Industrial communication
networks - Network and system
security
Part 2-1: Establishing an industrial automation and control system security program
ISO/IEC/IEEE
29148:2011
Systems and software engineering
-- Life cycle processes Requirements engineering
IEC 60870-5:2018 Telecontrol equipment and
systems Part 5: Transmission protocols - all parts
ISO 21500:2012 Guidance on project management Guidance on project management
Fig. 1. Levels of Collaboration [14].
470 Daniel Tokody et al. / Procedia Manufacturing 46 (2020) 467–474
4 Daniel Tokody et al. / Procedia Manufacturing 00 (2019) 000–000
3. Robot security features and risk reduction
The following safety functions support human - machine interaction. The robot movement can be stopped before
the operator enters the collaborative space. As a general rule, while the operator is not in the collaborative space, the
robot is not working collaboratively. When the operator enters the safety zone of the robot, the robot first reduces its
operation and then stops it. The robot starts working at maximum speed when the operator leaves the safety zone.
With manual control, the safety zone is flexible and depends on the speed of the job and the task that is manually
controlled by the operator. When implementing speed and distance monitoring, the cobot and operator can move in
the common area at the same time. Depending on the safety zone, the cobot reduces, stops or increases its speed in
response to the operator's reactions. After stopping, the cobot can restart automatically or manually as instructed by
the operator. The robotic arms are equipped with torque and force limits [8] to prevent the operator from
accidentally striking the front. [15]
Using collaborative robots reduces the risk to cause injuries to operators. An important aspect when designing
collaborative robots is the exterior design. This plays a big role in reducing risks (e.g. the edges are always
rounded). These robots can be mobilized and require little space. They are suitable for the operator and the robot to
work side by side without physical fencing. Communication links are compact in design, which means that wiring
harnesses are in most cases hidden. Collaborative robots allow the person working with them not to get tired of
monotonous work. This is due to the advanced sensor system of robots, which always gives priority to safety. [15]
[16]
4. The SUBIKO® research and development project goals
The goal of the development is to create a cooperative robotic system (integration of multiple cooperative robotic
arms and their environment / cooperative and non-cooperative machines and people). The system should meet that
meets the increased safety requirements, which reduces the risk of accidents between machines and people, while
maximizing the availability, reliability and operational, migration costs Maintainability of the system [6] is expected
to be low not only in mass production, but also in smaller varied series, flexibly without costly switching through
online risk analysis (used techniques: machine learning, deep learning [17] [18]) and determinable AI (Artificial
Intelligence). [19] [20] [21] [22] That means the project software elements Robot Interconnect Distributed
Scheduler and Robot Supervisor.
During development, the technologies required by the standards are applicable. If there are no such standards in
the particular field, the applicability of the technology for the development purpose is proven by experiments,
measurements and simulation. When designing parts of a cobot system, we carry out a risk analysis as required by
the standards, which we use to categorize which items fall into a risk category. For security-critical system
components, we apply security-critical hardware and software architecture and development methodology. [23]
During development, we determine the required safety integrity level (SIL) for each subsystem. The feature of the
cooperative robotics system that makes human interaction safer is the image processing system (human, obstacle
point cloud, workpiece positioning and orientation. human nervous system perception, robotic environment
recognition) and the services based on it greatly increase the flexible usability of the cobots. It is necessary to
determine which procedures can be integrated into the cobot system, which can be expanded later, and users can
upload their own unique image processing procedure to an image processing routine. [24]
Customized manufacturing of smaller varied series requires a high degree of flexibility and is expected to play a
role in the process. Maintaining safety when in the workplace requires physical measures and requires lengthy risk
analysis. Safe handling of such complexity is difficult due to the size of the large problem space. Full testing is not
possible. A general principle in achieving security is that complexity should be kept low. This is combined with
flexible operational needs, which are mutually exclusive. One solution to this is the systematic problem resolution,
in which we break down complex problems into smaller parts and then solve them by repeating simple rules several
times. During the project it should be examined whether the security tasks identified during the design of the cobot
system support the expected production flexibility with such a systematic breakdown. The security mode required
by the robot system depends on the environment. According to our preliminary research, we implement two main
security approaches (reliability systems) in integrating Fails-safe and Fail Operational in the operation of the cobot.
Daniel Tokody et al. / Procedia Manufacturing 46 (2020) 467–474 471
Daniel Tokody et al. / Procedia Manufacturing 00 (2019) 000–000 5
[3] The conflicting implementation of these two modes of operation does not allow the creation of a general purpose
cobot system. When designing reliability, we place great emphasis on integrating cyber security and integrating it
into the development process. To ensure the flexibility of the cobot system, we use AI (Artificial Intelligence). We
intend to prove the applicability of AIs in task-specific security environments through research and experimentation
during the project. The subprojects aim to increase security in a flexible layout and application environment. By
implementing a cobot system we enable the top-down decomposition of tasks. If multiple robotic arms or people,
and other equipment are required to work together on a job, the tasks are rescheduled online according to the current
operating conditions. The system is able to perform risk analyses necessary for its operation, and is also able to
detect potential hazards, alongside with their expected probability and the expected extent of damage event. By
comparing alternatives, the system has the capability of making predictions during operation to assist itself. These
security measures created by the cobot system during operation are implemented by the system. Based on the self-
generated emergency scenario, fail-safe or fail-operational mode will be implemented. Figure 2 illustrates the
security operation of a cooperative robot. [24]
5. Architecture and Systems Engineering Models and Methods to Manage Complex Systems
The following section details the adaptation of Scrum v Agile development methodologies in cobot system
development.
Scrum is a methodological framework used in software development. With this method, companies can easily
develop and support a product in a complex and dynamic environment [25]. Scrum is the answer to the rapid
evolution of technology and the rapid change in customer needs. The starting point of the method is experience. We
try and learn from our experiences and decide how to continue to develop. Scrum teams consist of 4-5 people in the
project. These teams control themselves. Team members work on a step-by-step basis. At specific stages, a new
product / component, a component or function that can be identified as an element or an improved version of an
earlier result product is created. These races are called sprints. The methodology will allow for continuous and
realistic development on a timely basis. [26] [27]
Splitting work into sprints is ideal because it allows the team to plan the workflow more realistically, as they can
see more precisely what should be done and how much time it will take. Therefore, the design process will be much
more predictable. Developments in short periods are also better for risk management. Shorter sub-tasks do not
require lengthy comprehensive planning and complex risk analysis at the same time. The results achieved by the
teams in the short term become visible to those involved in the development and to the customer. Obstacles to the
development process can also be tackled more quickly and easily. Short time intervals allow for more transparency
in team work and project management to intervene quickly for a project if needed. At the end of each sprint, the
results of the development can be presented to the customer. The project becomes more transparent, so the customer
can provide direct feedback that the team can use in the next sprint. This is how we develop the cobot system as a
product to meet the needs of the customer. [26] [27]
In the Agile methodology, individual and personal communication is more important than processes and tools,
working software is more important than comprehensive documentation, cooperation with the customer is more
important than contract negotiation, and responding to change is more important than rigid follow-up of plans.
Twelve principles have been established in the Agile methodology. The Agile mindset can also be useful in
development. The point of introducing agile methodology is to make larger projects more flexible and responsive to
change. Larger projects are less flexible. Projects typically use the waterfall model. In this case, you first have to go
through certain phases, which can be lengthy, thus making development work slow. Agile methodology changes
this. [28]
What is the difference between Scrum and Agile methodologies?
The idea behind Agile is that a project that wants to "survive" must be flexible and adaptable to change.
Development projects must be flexible enough to adapt to new technologies and changing customer needs.
An Agile mindset allows you to return flexibility to your project and react quickly to change. In an Agile
developer work environment, employees need to share knowledge, be creative, and find the solution themselves.
The initiative is no longer a task for leaders but for researchers and developers in the field.
How does Agile work in practice?
472 Daniel Tokody et al. / Procedia Manufacturing 46 (2020) 467–474
6 Daniel Tokody et al. / Procedia Manufacturing 00 (2019) 000–000
In a traditional work setting, we try to rule out change: we make a plan and try to stick to it as much as possible.
The starting point in agile thinking is that plans can change. It is not always possible to follow the same plan from
start to finish. The development goal is clear, but the path to it can change. Agile work is a never-ending
uninterrupted process of improvement. [29] [28]
When a cobot is developed, using agile methodology with scrum is strongly proposed. The development without
agile and scrum, but with the V model or waterfall (for example) would take much more time. It is very likely that
due to the long development time, the market or parameters will change. If the requirements are changing the cobot
needs to be modified accordingly. A simple V model or a waterfall would not be able to follow without starting the
design and development from the beginning and therefore losing development time.
Scrum can easily follow the changes between the release cycles of the project and adopt to the new requirement
minimizing the time cost. As mentioned before, the key feature in these methodologies is the quick tracking of
changes which makes them perfect for cobot system development.
6. Results and Discussion
The reviewed literature and the examined cobots are very different from the cobot system we have designed. As
part of the project, Robot Interconnect Distributed Scheduler and Robot Supervisor will carry out online risk
analysis using artificial intelligence. This increases the availability, speed and security of the robots. Cobots must
behave in a fail-safe or fail-operational manner, depending on the application. Current cobots use only one of these
methodologies; most operate in a fail-safe manner. Fail-operational mode mostly a requirement for robots applied in
the medical field. The difference between the two operating modes can be decided on the basis of risk analysis.
When stopping the cobot involves more risk than further running it, the fail-operational mode is expected. If
stopping the cobot is more likely to cause accidents (or material damage) then fail-safe operation is expected.
Illustratively, the aircraft's on-board systems cannot be stopped, because it would cause the aircraft to be
uncontrollable
The typical behaviour of the two modes of operation for sensor failure is as follows. The fail-safe sensor switches
to safety mode in the event of an error. Either it stops immediately, or the system moves to the designated safety
position at a safe speed. Sensor duplication is used to detect sporadic error by comparison. In the event of a fail-
operational failure the sensor switches to redundant or secondary (based on data from other sensors) detection mode
and continues to work in a safe mode (by reducing risks as much as necessary and possible / eg engine speed
limitation). The project's AI management system and extended sensor network will allow fail-safe and fail-
operational modes. This requires online and real-time risk analysis. The project will develop an AI network capable
of risk analysis and determination of the output of AI.
The schematic HL (High Level) decision process can be interpreted as a closed chain. The system makes the
decision based on the data collected by the senor network. This decision is split because it can affect more robots
(distributed decision making). [30] [31] Therefore, the decision has global and local levels. Both can make a security
decision, but local is a priority. The local may decide to accept the global security authorization decision. But this
decision can only be for a short period of validity and the current level of perceived worst-case risk should not
exceed a predetermined level of risk.
The Control module performs a risk analysis and, after making a decision, re-verifies the decision risk before
making a decision. After the outage, the system re-analyses the risk, waiting for the time constant of the system to
qualify the previous decision. The rating is shaped by two factors. The decision includes known and predicted data.
The rating is possible if the predicted data match. If the prediction was unsuccessful, it will be stored and will later
improve its own prediction model. The robot arm can make its own decisions, but it can only perform the functions
according to the standard ISO / TS 15066 standard.
Multiple structured or teaching AIs perform the same task, and then a statistic-based comparator chooses the right
decision. Then, we compare the AI decisions. All inputs and decisions, and also the consequences are stored per AI,
thus building a problem space-decision-efficiency database per AI. The system then applies a weighting to the
decisions and carries out a risk analysis for each decision. The internal schematic of Control is as follows. The
sensor performs a risk analysis based on network data, then actualises the system model and makes a decision. The
model tests the decision. Based on the model, it predicts the new condition and assesses its risk. If appropriate, the
Daniel Tokody et al. / Procedia Manufacturing 46 (2020) 467–474 473
Daniel Tokody et al. / Procedia Manufacturing 00 (2019) 000–000 7
decision will be ran in the rule database. If appropriate, the decision will be outputted; if not, then restrictions will
follow based on the rules. Automatic production layout recognition: with the help of unified messaging and object-
oriented communication, the system is able to find and communicate with ring-connected equipment. If the system
is in reconnaissance mode, the devices will move safely one by one and the system will use its sensors to determine
where they are moving. In this way, in addition to detecting the logical system, it is capable of physical assessment
and interconnecting the two. This will automatically record the production layout. The likelihood of image
recognition uncertainty is complemented by a risk assessment procedure that judges the results of several different
methods of image processing based on the robot's environment and operating condition. If there are several risk
assessment options at system startup or when the environment changes dramatically, the system will use the one
with the highest risk value. The system continually calibrates itself to assess the risk best suited to the environment.
7. Conclusion
The coordination of the two non-cooperative management systems (Fail-Safe, Fail-Operational) can be achieved
through online risk assessment We can create additional security information for ancillary systems (which is
required for risk analysis), and more detailed knowledge of the environment and support for automated task
breakdown and execution. By combining fail-safe and fail-operational tasks with current scientific knowledge,
robotic-controlled and human-based work sets up efficient workflow based on pre-designed data that eliminates the
risk of human and technical failure where one operating mode would not have been enough and it is only discovered
after the incident that the other operating mode could have avoided this accident. Understanding the scientific
principles or relationships to drive forward this project overcomes the professional preconceived notion of how a
previously unmatched security technology approach can improve efficiency in the manufacturing process and
facilitate human-robot collaboration. Choosing between Fail-Safe and Fail-Operational may present a risk analysis
for the scientific uncertainty that hinders the achievement of the set goals. A new concept to eliminate scientific
uncertainty could be the development of a cooperative robotic system that meets increased security requirements, a
project that can be deployed more rapidly and provides for faster reprogramming to increase the overall security of
cobots, analyse human behaviour, perform automated cooperative tasks, and create security artificial intelligence.
Acknowledgement
The research on which the publication is based has been carried out within the framework of the project entitled
“Low-load universal cooperative industrial robot system with innovative security solutions”. (Application number:
GINOP-2.1.2-8-1-4-16-2018-00492) This work was supported by NextTechnologies Ltd. Complex Systems
Research Institute.
References
[1] A. Rodić, M. Jovanovic, S. Popic, G. Mester, Scalable experimental platform for research, development and testing of networked robotic
systems in informationally structured environments: Experimental testbed station for wireless robot-sensor networks. In Symposium Series
on Computational Intelligence, IEEE SSCI 2011 - 2011 IEEE Workshop on Robotic Intelligence in Informationally Structured Space, 2011,
pp. 136–143.
[2] G. Mester, A. Rodić, Autonomous locomotion of humanoid robots in presence of mobile and immobile obstacles trajectory prediction, path
planning, control and simulation. Studies in Computational Intelligence. University of Szeged; 2009, pp. 279–293.
[3] P. Francesco, G. G. Paolo, AURA: An Example of Collaborative Robot for Automotive and General Industry Applications. Procedia
Manuf., 11 (2017) 338–345.
[4] J. Cosic, P. Curkovic, J. Kasac, J. Stepanic, Interpreting Development of Unmanned Aerial Vehicles using Systems Thinking. Interdiscip.
Descr. Complex Syst. 11:1 (2013) 143–152.
[5] L. Moldovan, A. Gligor, Global trends in manufacturing technologies which create a path to tomorrow’s innovations. Procedia Manuf.; 22
(2018) 1–3.
[6] D. Schröter, P. Jaschewski, B. Kuhrke, A. Verl, Methodology to Identify Applications for Collaborative Robots in Powertrain Assembly.
Procedia CIRP. 55 (2016) 2–17.
[7] Z. Illési, Robot Incident Investigation: A Computer Forensic Approach, in Proceedings of ARES’14 : Workshop on Application of Robotics
for Enhanced Security. 2014; 30–34.
474 Daniel Tokody et al. / Procedia Manufacturing 46 (2020) 467–474
8 Daniel Tokody et al. / Procedia Manufacturing 00 (2019) 000–000
[8] B. Vemula, B. Matthias, A. Ahmad, A design metric for safety assessment of industrial robot design suitable for power- and force-limited
collaborative operation. Int. J. Intell. Robot. Appl. 2:2 (2018) 226–234.
[9] L. Seminara, P. Gastaldo, S. J. Watt, K. F. Valyear, F. Zuher, F. Mastrogiovanni, Active Haptic Perception in Robots: A Revie. Front.
Neurorobot. 13 (2019) 1–20.
[10] R. Leenes, E. Palmerini, B.-J. Koops, A. Bertolini, P. Salvini, F. Lucivero, Regulatory challenges of robotics: some guidelines for addressing
legal and ethical issues. Law, Innov. Technol. 9:1 (2017) 1–44.
[11] S. R. Hirshorn, NASA System Engineering Handbook Revision 2., Washington: NASA; 2016, p. 297.
[12] ISO/TS 15066:2016: Technical Specification ISO / TS Robots and robotic devices — ISO/TS 15066:2016. 2016.
[13] T. Dokody, G. Schuster, I2 - Intelligent Infrastructure. In Fifth International Scientific Videoconference of Scientists and PhD. students or
candidates, 2015; 121–128.
[14] A. Kolbeinsson, E. Lagerstedt, J. Lindblom, Classification of collaboration levels for human-robot cooperation in manufacturing. Adv.
Transdiscipl. Eng. 8 (2018) 151–156.
[15] P. Wadekar, V. Gopinath, K. Johansen, Safe Layout Design and Evaluation of a Human-Robot Collaborative Application Cell through Risk
Assessment - A Computer Aided Approach. Procedia Manuf., 25 (2018) 602–611.
[16] J. Guiochet, M. Machin, H. Waeselynck, Safety-critical advanced robots: A survey. Rob. Auton. Syst. 94 (2017) 43–52.
[17] J. E. Diekmann, Risk analysis: lessons from artificial intelligence. Int. J. Proj. Manag. 10:2 (1992) 75–80.
[18] X. Xie, et al. Risk prediction and factors risk analysis based on IFOA-GRNN and apriori algorithms: Application of artificial intelligence in
accident prevention. Process Saf. Environ. Prot. 122 (2019) 169–184.
[19] S. T. Brassai, B. Iantovics, C. Enachescu, Artificial Intelligence in the path planning optimization of mobile agent navigation. Procedia
Economics and Finance. 2012; 3, 243–250.
[20] L. B. Iantovics, M. Dehmer, F. Emmert-Streib, MetrIntSimil-an accurate and robust metric for comparison of similarity in intelligence of
any number of cooperative multiagent systems. Symmetry (Basel). 2018; 10:2, 48.
[21] L. B. Iantovics, A. Gligor, M. A. Niazi, A. I. Biro, S. M. Szilagyi, D. Tokody, Review of Recent Trends in Measuring the Computing
Systems Intelligence. Broad Res. Artif. Intell. Neurosci. 2018; 9:2, 77–94.
[22] F. Flammini, Artificial Intelligence (AI) Applicata Agli Autonomous Systems. Roma: Centro Militare di Studi Strategici Dipartimento
Relazioni Internazionali, 2018.
[23] Z. Stojanov, J. Stojanov, D. Dobrilovic, Knowledge discovery and systematization through thematic analysis in software process assessment
project. In 2015 IEEE 13th International Symposium on Intelligent Systems and Informatics (SISY), 2015, 25–30.
[24] ‘Subiko project’, 2019. [Online]. Available: http://www.subiko.hu/projekt/. Accessed: 28-May-2019
[25] D. Tokody, J. Papp, L B. Iantovics, F. Flammini, Complex, Resilient and Smart Systems. In Resilience of Cyber-Physical Systems, F.
Flammini, Ed. Springer International Publishing, 2019, 3–24.
[26] M. Tytkowska, A. Werner, M. Bach, Project Management in the Scrum Methodology. 2015, 483–492.
[27] G. K. Hanssen, T. Stalhane, T. Myklebust, SafeScrum – Agile Development of Safety-Critical Software. Cham: Springer Nature Switzerland
AG; 2018, 233.
[28] T. Myklebust, T. Stalhane, The Agile Safety Case. Cham: Springer Nature Switzerland AG; 2018, 242.
[29] G. Bakker, Scrum és agile: mi a különbség közöttük és mire jók?. 2018. [Online]. Available: https://blog.topdesk.com/hu/scrum-es-agile.
Accessed: 06-Jun-2019
[30] A. Gligor, C D. Dumitru, Agents-Based Distributed Processes Control Systems. In The 5thEdition of theInterdisciplinarity in Engineering
International Conference “Petru Maior” University of Tîrgu Mureş. 2011; pp. 121–125.
[31] R. Pető, The „ Prophet ” decision support system. Procedia Manuf., 22 (2018) 1023–1030.
... With the aim of combining human abilities and cognitive capacities with robots' advantages in automation systems, it is crucial that they can share their workspace in safe conditions [1,2,10,[12][13][14][15][16][17][18][19][20], enabling their interaction. This interaction has been described by many authors [10,11,19,[21][22][23]. ...
... With the aim of combining human abilities and cognitive capacities with robots' advantages in automation systems, it is crucial that they can share their workspace in safe conditions [1,2,10,[12][13][14][15][16][17][18][19][20], enabling their interaction. This interaction has been described by many authors [10,11,19,[21][22][23]. Moreover, the ISO/TS 15066 standard defines to the following five interactions ( mance, repeatability, and a high adaptive capacity to perform different tasks [1]. ...
... Ho ever, their characteristics depend on conception and assembly elements such as the t [6],the tasks, the fixation on the work area [8], and their connection with other machin among others. With the aim of combining human abilities and cognitive capacities with robots' vantages in automation systems, it is crucial that they can share their workspace in s conditions [1,2,10,[12][13][14][15][16][17][18][19][20], enabling their interaction. This interaction has been descri by many authors [10,11,19,[21][22][23]. ...
Article
Full-text available
The coordination of two collaborative robots to handle and hold huge parts is the main topic of this research. This study shows how flexible systems may accommodate large-volume components while situating components with a displacement precision between robots of no more than 10 mm into the parts, with the assistance of a single operator. The robots must be able to keep the parts in place while coordinating their movements to handle the parts and reducing external stressors. This paper suggests using collaborative robots to integrate flexible tools for adaptability to various elements in order to accomplish this goal without endangering the operators. The software architecture is described in full in this paper, including machine states to choose task executions, robot referencing in the workspace, remote monitoring via the digital twin, generation paths, and distributed control using a high-level controller (HLC).
... e study proceeded as follows. e systematic literature review methodology is explained; the research process follows the protocol suggested in [1,27,28]. e search results as per the research questions and their quality assessment were done in the later sections. ...
... Articles are reviewed individually for relevance concerning abstracts and content. is filtration process aided in identifying the existing literature that can help to answer the research questions. e content of the article and the suitability of the article are verified with the help of literary works as discussed by Kitchenham et al. [27]. e graphs in Figures 6 and 7 explain the relative proportion of articles drawn from various renowned digital repositories for the research questions RQ1 and RQ2 individually. ...
Article
Full-text available
Artificial intelligence (AI) has become omnipotent with its variety of applications and advantages. Considering the other side of the coin, the eruption of technology has created situations that need more caution about the safety and security of data and systems at all levels. Thus, to hedge against the growing threats of cybersecurity, the need for a robust AI platform supported by machine learning and other supportive technologies is well recognized by organizations. AI is a much sought-after topic, and there is extolling literature available in repositories. Hence, a systematic arrangement of the literature that can help identify the right AI platform that can provide identity governance and access control is the need of the hour. Having this background, the present study is commissioned a Systematic Literature Review (SLR) to accomplish the necessity. Literature related to AI and Identity and Access Management (IAM) is collected from renowned peer-reviewed digital libraries for systematic analysis and assessment purposes using the systematic review guidelines. Thus, the final list of articles relevant to the framed research questions related to the study topic is fetched and is reviewed thoroughly. For the proposed systematic research work, the literature reported during the period ranging from 2016 to 2021 (a portion of 2021 is included) is analyzed and a total of 43 papers were depicted more relevant to the selected research domain. These articles were accumulated from ProQuest, Scopus, Taylor & Franics, Science Direct, and Wiley online repositories. The article's contribution can supplement the AI-based IAM information and steer the entities of diverse sectors concerning seamless implementation. Appropriate suggestions are proposed to encourage research work in the required fields.
Article
Full-text available
This article attempts to review papers on power assist rehabilitation robots, human motion intention, control laws, and estimation of power assist rehabilitation robots based on human motion intention in recent years. This paper presents the various ways in which human motion intention in rehabilitation can be estimated. This paper also elaborates on the control laws for the estimation of motion intention of the power assist rehabilitation robot. From the review, it has been found that the motion intention estimation method includes: Artificial Intelligence-based motion intention and Model-based motion intention estimation. The controllers include hybrid force/position control, EMG control, and adaptive control. Furthermore, Artificial Intelligence based motion intention estimation can be subdivided into Electromyography (EMG), Surface Electromyography (SEMG), Extreme Learning Machine (ELM), and Electromyography-based Admittance Control (EAC). Also, Model-based motion intention estimation can be subdivided into Impedance and Admittance control interaction. Having reviewed several papers, EAC and ELM are proposed for efficient motion intention estimation under artificial-based motion intention. In future works, Impedance and Admittance control methods are suggested under model-based motion intention for efficient estimation of motion intention of power assist rehabilitation robot. In addition, hybrid force/position control and adaptive control are suggested for the selection of control laws. The findings of this review paper can be used for developing an efficient power assist rehabilitation robot with motion intention to aid people with lower or upper limb impairment.
Article
Full-text available
The term "collaborative robot" is commonly known as Cobot, which refers to a partnership between a robot and a human. Aside from providing physical contact between a robot and a person on the same production line simultaneously, the Cobot is designed as user-friendly. They enable operators to respond immediately to work done by the robot based on the company's urgent needs. This paper aims to explore the potential of Cobots in manufacturing. Cobots are widely employed in various industries such as life science, automotive, manufacturing, electronics, aerospace, packaging, plastics, and healthcare. For many of these businesses, the capacity to maintain a lucrative man-machine shared workplace can provide a considerable competitive edge. Cobots are simple to use while being dependable, safe, and precise. A literature review was carried out from the database from ScienceDirect, Scopus, Google Scholar, ResearchGate and other research platforms on the keyword “Cobots” or “Collaborative robots” for manufacturing. The Paper briefly discusses and provides the capabilities of this technology in manufacturing. Cobots are programmed to do crucial things such as handling poisonous substances, from putting screws on a vehicle body to cooking a meal, etc. Human operators can readily control this technology remotely and perform dangerous jobs. This paper's overview of Cobots and how it is differentiated from Robot is briefly described. The typical Features, Capabilities, Collaboration & Industrial Scenarios with Cobots are also discussed briefly. Further, the study identified and discussed the significant applications of Cobots for manufacturing. Cobots are utilised in several methods and a wide range of application areas. These elevate manufacturing and other operations to new heights. They also collaborate with humans to balance the demand for safety and the need for flexibility and efficiency.
Chapter
While artificial intelligence (AI) experienced an increasing interest in industry during the past decade, the true potential and applicability of AI for automotive original equipment manufacturers (OEMs) and suppliers in real-world scenarios have not been clearly understood. Most applications of AI focus on the development of connected and autonomous cars, rather than the optimisation of automotive operations and manufacturing processes. This work, therefore, bridged this gap and shed light on the topic of AI in the context of automotive manufacturing and Industry 4.0. It aimed to promote understanding and provide up-to-date insights on specific models and methods of AI, applications that have been achieved with best practices as well as the problems that were encountered, underpinned with possible future prospects. A systematic literature review approach was adopted to ensure broad and thorough coverage of current knowledge and the identification of relevant literature on the topic. The literature search was confined to papers that were published from 2015 onwards using the databases of IEEE and ScienceDirect as primary sources, with a three-keyword search phrase to narrow down the results and increase specificity. A total of 359 papers were identified and subsequently screened for eligibility, of which 84 papers were selected for quantitative and 79 papers for qualitative analysis. The results of the quantitative analysis confirmed that the topic has markedly increased in significance, with a mere 3 papers published in 2015 and 33 papers in 2021. The majority of papers dealt with solving problems in production (39.29%), quality (35.71%) and assembly (16.67%), whereas supply chain (5.95%) and business intelligence (2.38%) were inadequately represented. The results of the qualitative analysis revealed that machine learning methods dominate current research and automotive applications, with neural networks as the most used out of more than 70 identified models. The industrial applicability was confirmed by many use cases including quality inspection, robot assembly, human–robot collaboration, material demand prediction or AI-enabled manufacturing decision making. The problems of such applications were mainly attributed to data availability and quality, model development and gaps in simulation, system integration, the complexity of automotive processes, the physical conditions of the system environment and dynamic change. For industrial applications it is thus recommended to further optimise AI methods and models, enabling a wider system integration by harvesting the potential of big data and both edge and cloud computing.KeywordsArtificial intelligenceAI models and methodsAutomotive manufacturingSystematic literature review
Article
From 2011 onwards, thanks to the Industry 4.0 concept referring to the fourth industrial revolution, different technologies have emerged with the main objective of achieving increased productivity. One of the main technologies related to Industry 4.0 is collaborative robots, also known as cobots, whose main feature is an intelligent real-time labor nexus between humans and robots respecting safety standards. Therefore, the aim of this research work is to identify, analyze and map the technological profile of collaborative robots through patents. The results obtained show that the number of patents in recent years has grown exponentially since 2016. The largest number of inventions are concentrated in China, led by different universities, research centers and companies. Other important countries, in terms of development, are Japan and Germany, however, the inventions are mainly led by one company, Fanuc Corporation and Kuka Roboter GmbH. Likewise, it is clear that the technological advances are mainly directed toward technological developments linked to manipulators and hand tools, specifically to safety accessories and program-controlled manipulators. The results of the research provide the community involved in the research and development of collaborative robotics with a solid conceptual review, as well as a current snapshot of the temporal evolution and specialization in the technological field of inventions, obtaining intelligence information about Collaborative Robot technology to support R&D strategies.
Article
Full-text available
Human behavior, despite its complexity, follows structured principles. Better understanding the underlying concepts of human behavior in an industrial setting can help us to create more reliable and effective collaborative robotic environments. In factories, robots can interact with humans in unseen situations through sensing, processing , and predicting human behaviors. This work summarizes the efforts to express human behavior in industrial human-robot interactions (HRI) in manufacturing. For this purpose, all papers related to HRI in production systems, additive manufacturing, agent behavior, human cognition, cloud robotics, cooperative Markov decision process, multi-robot design, and collaborative robots are reviewed in this work. The discussion includes the existing methods, research gaps, and future research directions for envisioning a safer yet more efficient HRI practice. In detail, we discuss current research about human-centered HRI and robot-centered HRI based on the focus areas such as factorial analysis, predictive analysis, and control structures for social and/or industrial/manufacturing robots. In the last part, based on our findings, we report major limitations of the existing literature and propose future research directions such as cognitive modeling, perception development, interaction design, sensor-based control, and social effects in manufacturing.
Article
Full-text available
Industry 4.0 aims to support the factory of the future, involving increased use of information systems and new ways of using automation, such as collaboration where a robot and a human share work on a single task. We propose a classification of collaboration levels for Human-Robot collaboration (HRC) in manufacturing that we call levels of collaboration (LoC), formed to provide a conceptual model conducive to the design of assembly lines incorporating HRC. This paper aims to provide a more theoretical foundation for such a tool based on relevant theories from cognitive science and other perspectives of human-technology interaction, strengthening the validity and scientific rigour of the envisioned LoC tool. The main contributions consist of a theoretical grounding to motivate the transition from automation to collaboration, which are intended to facilitate expanding the LoC classification to support HRC, as well as an initial visualization of the LoC approach. Future work includes fully defining the LoC classification as well as operatio-nalizing functionally different cooperation types. We conclude that collaboration is a means to an end, so collaboration is not entered for its own sake, and that collaboration differs fundamentally from more commonly used views where automation is the focus. ARTICLE HISTORY
Article
Full-text available
Risk prediction of disasters is one of the most effective ways to prevent accidents. To solve the problems in multi-factor complex disaster prediction, this paper proposes a new method for risk prediction and factorial risk analysis. Coal and gas outburst accidents were selected as research objects. First, a new coal and gas outburst prediction model was established that consists of 4 levels and 14 factors. Then, the Improved Fruit Fly Optimization Algorithm (IFOA) and the General Regression Neural Network (GRNN) algorithm were combined to establish the IFOA-GRNN prediction model. After that, the sensitivity analysis method was applied to the analysis of the sensitive factors of coal and gas outbursts. Finally, an apriori algorithm was used to mine the disaster information. The method proposed in this paper was applied to the Pingdingshan No. 8 Min. The application results show that the IFOA-GRNN algorithm proposed in this paper has an accuracy rate of 100% for the prediction of accident risk levels. Compared with the Back Propagation (BP), GRNN and FOA-GRNN algorithms, IFOA-GRNN has the characteristics of a smaller prediction error, higher stability and faster convergence. The sensitivity analysis method can judge the sensitive factors of coal and gas outbursts without knowing the mechanisms of the accident. The a priori algorithm can perform good data mining on the combination of high frequency factors leading to accidents and the relationships between the coal and gas outburst levels and factors. The data mining results are very helpful for the prevention and management of coal and gas outbursts.
Article
Full-text available
Risk assessment is a sequential process which requires understanding the hazards and analyzing risks associated, to determine required safety measures, like safeguarding to mitigate the risks to an acceptable level. In this article a thorough task-based risk assessment process is conducted in the early stage of layout design and building of a collaborative cell for sealing application performed in aircraft industries using a medium sized industrial robot system integrated with safety control functions. This article will also discuss how simulation could contribute in eliminating the threats as required by the safety standards before investing in equipment for collaborative cell layout.
Article
Full-text available
Many difficult problems, from the philosophy of computation point of view, could require computing systems that have some kind of intelligence in order to be solved. Recently, we have seen a large number of artificial intelligent systems used in a number of scientific, technical and social domains. Usage of such an approach often has a focus on healthcare. These systems can provide solutions to a very large set of problems such as, but not limited to: elder patient care; medical diagnosis; medical decision support; out-of-hospital emergency care; drug classification among others. A recent key focus is that most of these developed intelligent systems are agent-based approaches, or in other words, they can be considered as agent-based intelligent systems (ABISs). ABISs are formally based on a set of interacting intelligent agents (IAs) in addition to the use of intelligent cooperative approaches namely forming intelligent cooperative multiagent systems (ICMASs). The main direction of study consists in the possibility to measure the artificial systems intelligence, frequently called machine intelligence quotient (MIQ). Recently, we performed some research related to the measuring of the machine intelligence. There is presented a comprehensive review of the scientific literature related to the measuring of the MIQ. We consider that the measuring of the machine intelligence is very actual and important, which could allow the differentiation of ABISs based on their intelligence, choosing of the agent-based systems able to solve the most intelligently specific problems. As the main conclusion of the performed study, we mention that cannot be given a unanimous definition of the ABISs intelligence. Even if the machine intelligence cannot be defined, it could be measured. We discuss this affirmation more in-depth in the paper. This is similar to the human intelligence that is not understood very well but can be measured using human intelligence tests.
Article
Full-text available
This research presents a novel design metric based on maximum power flux density for the assessment of the severity of a transient physical contact between a robot manipulator and a human body region. Such incidental transient contact can occur in the course of a collaborative application of the power- and force-limiting type. The proposed metric is intended for the design and development of the robot manipulator as well as for the design of manufacturing applications. Such safety metric can also aid in controlling the robot’s speeds during manufacturing operations by carrying out rapid risk assessments of impending collisions that could arise due to the proximity to the human co-worker. Furthermore, this study contributes by expressing the physical impact between the robot and the human body region as a linear spring-damper model. The influence of the restitution coefficient and the elasticity of the human tissues on the contact duration and contact area during the collision is analysed. With the demonstrated analysis model, the dependence of the power flux density with respect to the robot’s effective mass, speed, and geometrical and damping coefficients during the human-industrial robot manipulator collision process is investigated.
Article
Full-text available
The prevention, countering, and elimination of disaster situations and crimes require the coordinated work of special agencies and service providers. The Prophet is a map-based system with the objective to plan and efficiently coordinate the work processes of defence, law enforcement, disaster management, civil defence, and public services. The present article provides a brief description of the structure of the map system meeting the criteria.
Article
Full-text available
Intelligent cooperative multiagent systems are applied for solving a large range of real-life problems, including in domains like biology and healthcare. There are very few metrics able to make an effective smeasure of the machine intelligence quotient. The most important drawbacks of the designed metrics presented in the scientific literature consist in the limitation in universality, accuracy, and robustness. In this paper, we propose a novel universal metric called MetrIntSimil capable of making an accurate and robust symmetric comparison of the similarity in intelligence of any number of cooperative multiagent systems specialized in difficult problem solving. The universality is an important necessary property based on the large variety of designed intelligent systems. MetrIntSimil makes a comparison by taking into consideration the variability in intelligence in the problem solving of the compared cooperativemultiagent systems. It allows a classification of the cooperativemultiagent systems based on their similarity in intelligence. A cooperative multiagent system has variability in the problem solving intelligence, and it can manifest lower or higher intelligence in different problem solving tasks. More cooperative multiagent systems with similar intelligence can be included in the same class. For the evaluation of the proposed metric, we conducted a case study for more intelligent cooperative multiagent systems composed of simple computing agents applied for solving the Symmetric Travelling Salesman Problem (STSP) that is a class of NP-hard problems. STSP is the problem of finding the shortest Hamiltonian cycle/tour in a weighted undirected graph that does not have loops or multiple edges. The distance between two cities is the same in each opposite direction. Two classes of similar intelligence denoted IntClassA and IntClassB were identified. The experimental results show that the agent belonging to IntClassA intelligence class is less intelligent than the agents that belong to the IntClassB intelligence class.
Article
Full-text available
AURA is the first high payload collaborative robot ever designed. The concept was developed in collaboration with the most famous Italian Universities and the technology (ISO certified) is based on redundant communication between a combination of sensors placed on the external surface, able to detect human proximity. AURA is the first step of Man-Machine collaboration, one of the Industry 4.0 main features.
Book
The study of artificial intelligence applied to autonomous systems has in recent years aroused growing interest at the international level, and it is expected that this interest will continue to grow in the coming years [34]. It is a fairly well known fact that in the past many technologies now used in the civil field have seen the light, more or less secretly, in the military sector. Consider, for example, the so-called ARPANET, developed by the US defense department, which anticipated the modern Internet, but also algorithms for data encryption, thermal cameras, and many other commonly used technologies. Today the scenario has partly changed, shifting the leadership of innovation towards other domains, since there is a considerable boost to the technological development in the civil field with the advance of connected society paradigms like Smart-City and Industry 4.0. One example is related to the self-driving vehicles, born in the military sector, which are developing more rapidly in the civil sphere with the attractive self-driving cars. It is therefore important to transfer enabling technologies from one domain to another (cross-fertilization) and to draw appropriately from the outside (open innovation). This is achieved through studies and researches such as the one addressed by this monograph. The objective of this study is to analyze the principles, the basic methodologies and the operational tools of artificial intelligence applied to autonomous systems, at the modeling and technology level, in order to replace human-controlled vehicles with autonomous or semi-autonomous vehicles (e.g. drones) in high-risk operating environments, as well as to reduce human errors and to speed up response times, for example in operations command and control centers. The study presents an overview of the information fusion approaches to enable artificial cognition, mentioning several relevant applications in the military field, already at an advanced phase of development or even at an embryonic level. These approaches can be used to strengthen weapon systems and defense means, with greater ability to adapt to the operational context for the dynamic management of uncertainties and unforeseen events, as well as for experiential evolution and learning. Future applications include not only self-driving vehicles and smart weapons, but also the strengthening of soldiers through prosthetics and exoskeletons. Many of the future projections have been formalized by the working group on Symbiotic Autonomous Systems – which the writer is a member of – of the Institute of Electrical and Electronics Engineers (IEEE), enclosed in a special White Paper [34]. The present study addresses the impact of the Artificial Intelligence (AI) on the use of the military instrument when this technology will be applied to military assets and weapon systems, taking into account the different declinations of AI, including: • deterministic (semi)autonomous systems implemented through Boolean logical operators (eg Event Trees); • (semi)Autonomous systems based on probabilistic / stochastic models for the representation of knowledge and inference (eg Bayesian Networks); • (semi)Autonomous systems based on trained artificial neuronal models (ANN, Artificial Neural Networks). These approaches are based on different models of machine learning, which can be supervised or not. They apply to classification and clustering approaches in modern data analysis approaches, particularly in the presence of large amounts of information (big data analytics). This study distinguishes between semi-autonomous AI models, which require the confirmation of decisions by human operators (DSS, Decision Support Systems), and complete autonomy, which presents predictability problems impacting the verification and validation process and therefore system safety. These are the cases in which the aforementioned ethical, procedural, normative and legal implications are more relevant [1]. The introduction of autonomous systems equipped with artificial intelligence involves transformations also at the level of military logistics, which can be interpreted in two directions. On the one hand, it is necessary to plan the procurement of enabling technologies, the so-called deployable systems based on secure wireless networks, and the updating of systems to support complete digitalisation, which is an essential pre-requisite for the adoption of the instrument. The other side of the coin is the use of a higher level of automation in military logistics, supported by the AI. Here we can mention the automatic multi-objective optimization algorithms for decision support (eg genetic and evolutionary programming), the computation of the most efficient paths (in terms of time, energy, etc.), the dynamic definition of optimization priorities, as well as aspects of resilience through automatic re-planning of the route in the event of interruptions on the predefined trajectory. For all that has been said so far, it is clear that the development of the AI will have consequences on the future organization of the armed forces, both for the conduct of the operations and for the structure and numbers of the defense sector. As in other areas subject to automation through the use of new digital technologies, even in the military one the human role of decision supervision, feedback and control of high-level operations will remain decisive for many years. At the same time, however, the need for training and specialization in line with the complete computerization will arise, with significant impacts in terms of information security (or cybersecurity), which will require increasingly specific skills. The fact that complete autonomy would be possible in the event of unavailability of personnel in control centers implies not only a higher level of security, but also the possibility of reducing organizational redundancies by dedicating resources to different and more specialized tasks. As already underlined, there are significant ethical and legal implications related to future decision-making processes for the choice of using force through a weapon system governed by an artificial intelligence, potentially endowed with a high level of autonomy. It is therefore essential to define clear and shared limitations and conditions of autonomy for the verifiability and traceability of the decision-making process. In particular, in order to govern decision-making and prevent ambiguities, it is essential to apply the well-known RACI (Responsible Accountable Consulted Informed) paradigm, which defines for each action who is responsible for its implementation, who is associated with its administrative / legal responsibility, who will have to be consulted for further information and possible approval, and finally who will have to be simply, but obligatorily, informed. All aspects related to international safety certifications that regulate design, development and verification of systems whose malfunctions can impact on the safety of people are also essential. Many of the current reference standards are no longer adequate if we consider the current and anticipated evolution of AI, and therefore they will have to be adjusted accordingly.