A Review on Integrating Autonomy Into System of Systems: Challenges and Research Directions
Artificial intelligence and machine learning (AI/ML) technologies convert conventional engineered systems into autonomous systems that are capable of performing tasks in their operational environment with limited to no human involvement. These technologies can reduce demands for human workload while...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Open Journal of Systems Engineering |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10669760/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850226661213077504 |
|---|---|
| author | Mohammadreza Torkjazi Ali K. Raz |
| author_facet | Mohammadreza Torkjazi Ali K. Raz |
| author_sort | Mohammadreza Torkjazi |
| collection | DOAJ |
| description | Artificial intelligence and machine learning (AI/ML) technologies convert conventional engineered systems into autonomous systems that are capable of performing tasks in their operational environment with limited to no human involvement. These technologies can reduce demands for human workload while enabling a suite of new capabilities such as autonomous vehicles and smart cities. However, a major challenge is the integration of these autonomous systems into a system of systems (SoSs), essentially resulting in a system of autonomous systems (SoASs). SoASs are fraught with new challenges that compound issues from the founding domains of SoS, AI/ML, and autonomous systems. To understand the new set of challenges for SoAS, this article conducts an extensive review of both theoretical and application-based literature published in the founding domains. The goal is to examine how individual challenges in each domain intersect and exacerbate when multiple independent systems with AI/ML are integrated into an SoAS. A particular emphasis is placed on highlighting how interactions across these domains manifest at the SoAS level. As a result, four overarching challenges for SoASs are identified that must be addressed by systems engineers to ensure a successful realization of the SoAS in the future: 1) SoAS foundation; 2) emergence, safety, and performance; 3) architecture and integration; and 4) test and evaluation. Each challenge is comprehensively examined in the three founding domains by discussing domain-specific state-of-the-art methods and tools that different engineering disciplines have proposed to address. For each challenge, we also investigated how the existing tools and methods apply to addressing the challenge for SoAS and highlighted the remaining gaps that still need to be addressed. Furthermore, this article identifies systems engineering research needs for improving SoAS foundations, analysis of autonomy impacts, and enabling SoAS architecture, integration, and evaluation methods. Conducting research studies in these fields will improve systems engineering practices for a successful and effective realization of SoAS. |
| format | Article |
| id | doaj-art-fd2e448df97c4373a41f61d4f5fd0858 |
| institution | OA Journals |
| issn | 2771-9987 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Systems Engineering |
| spelling | doaj-art-fd2e448df97c4373a41f61d4f5fd08582025-08-20T02:05:01ZengIEEEIEEE Open Journal of Systems Engineering2771-99872024-01-01215717810.1109/OJSE.2024.345603710669760A Review on Integrating Autonomy Into System of Systems: Challenges and Research DirectionsMohammadreza Torkjazi0https://orcid.org/0000-0001-8866-3009Ali K. Raz1https://orcid.org/0000-0003-2562-1631Department of Systems Engineering and Operations Research, George Mason University, Fairfax, VA, USADepartment of Systems Engineering and Operations Research, George Mason University, Fairfax, VA, USAArtificial intelligence and machine learning (AI/ML) technologies convert conventional engineered systems into autonomous systems that are capable of performing tasks in their operational environment with limited to no human involvement. These technologies can reduce demands for human workload while enabling a suite of new capabilities such as autonomous vehicles and smart cities. However, a major challenge is the integration of these autonomous systems into a system of systems (SoSs), essentially resulting in a system of autonomous systems (SoASs). SoASs are fraught with new challenges that compound issues from the founding domains of SoS, AI/ML, and autonomous systems. To understand the new set of challenges for SoAS, this article conducts an extensive review of both theoretical and application-based literature published in the founding domains. The goal is to examine how individual challenges in each domain intersect and exacerbate when multiple independent systems with AI/ML are integrated into an SoAS. A particular emphasis is placed on highlighting how interactions across these domains manifest at the SoAS level. As a result, four overarching challenges for SoASs are identified that must be addressed by systems engineers to ensure a successful realization of the SoAS in the future: 1) SoAS foundation; 2) emergence, safety, and performance; 3) architecture and integration; and 4) test and evaluation. Each challenge is comprehensively examined in the three founding domains by discussing domain-specific state-of-the-art methods and tools that different engineering disciplines have proposed to address. For each challenge, we also investigated how the existing tools and methods apply to addressing the challenge for SoAS and highlighted the remaining gaps that still need to be addressed. Furthermore, this article identifies systems engineering research needs for improving SoAS foundations, analysis of autonomy impacts, and enabling SoAS architecture, integration, and evaluation methods. Conducting research studies in these fields will improve systems engineering practices for a successful and effective realization of SoAS.https://ieeexplore.ieee.org/document/10669760/Autonomy integrationSoS challengesSoS engineeringsystem of autonomous systems (SoASs)system of systems (SoSs) |
| spellingShingle | Mohammadreza Torkjazi Ali K. Raz A Review on Integrating Autonomy Into System of Systems: Challenges and Research Directions IEEE Open Journal of Systems Engineering Autonomy integration SoS challenges SoS engineering system of autonomous systems (SoASs) system of systems (SoSs) |
| title | A Review on Integrating Autonomy Into System of Systems: Challenges and Research Directions |
| title_full | A Review on Integrating Autonomy Into System of Systems: Challenges and Research Directions |
| title_fullStr | A Review on Integrating Autonomy Into System of Systems: Challenges and Research Directions |
| title_full_unstemmed | A Review on Integrating Autonomy Into System of Systems: Challenges and Research Directions |
| title_short | A Review on Integrating Autonomy Into System of Systems: Challenges and Research Directions |
| title_sort | review on integrating autonomy into system of systems challenges and research directions |
| topic | Autonomy integration SoS challenges SoS engineering system of autonomous systems (SoASs) system of systems (SoSs) |
| url | https://ieeexplore.ieee.org/document/10669760/ |
| work_keys_str_mv | AT mohammadrezatorkjazi areviewonintegratingautonomyintosystemofsystemschallengesandresearchdirections AT alikraz areviewonintegratingautonomyintosystemofsystemschallengesandresearchdirections AT mohammadrezatorkjazi reviewonintegratingautonomyintosystemofsystemschallengesandresearchdirections AT alikraz reviewonintegratingautonomyintosystemofsystemschallengesandresearchdirections |