Content deleted Content added
Khanhlinh.gs (talk | contribs) |
|||
Line 59:
This is related to '''autonomous foraging''', a concern within the sciences of [[behavioral ecology]], [[social anthropology]], and [[human behavioral ecology]]; as well as [[robot]]ics, [[artificial intelligence]], and [[artificial life]].<ref>{{cite book|author = Kagan E., Ben-Gal, I., (2015)|format = PDF|title = Search and Foraging: Individual Motion and Swarm Dynamics (268 Pages) | date=23 June 2015 |url = https://www.amazon.com/Search-Foraging-Individual-Motion-Dynamics-ebook/dp/B010ACWAXC?ie=UTF8&*Version*=1&*entries*=0|publisher = CRC Press, Taylor and Francis }}</ref>
'''Systemic robustness and real-world brittleness'''
Autonomous robots remain highly vulnerable to unexpected changes in real-world environments. Even minor variations like a sudden beam of sunlight disrupting vision systems or unanticipated terrain irregularities can cause entire systems to fail<ref>{{Cite news |last=Brondmo |first=Hans Peter |title=Inside Google’s 7-Year Mission to Give AI a Robot Body |url=https://www.wired.com/story/inside-google-mission-to-give-ai-robot-body/ |access-date=2025-08-25 |work=Wired |language=en-US |issn=1059-1028}}</ref>. This brittleness stems from robotics being an inherently '''systems problem''', where a deficiency in any module (perception, planning, actuation) can compromise the whole robot.
'''Open-world scene understanding'''
Robots often depend on datasets captured under controlled conditions, limiting their ability to generalize to novel, dynamic real-world scenarios. They struggle with unknown objects, occlusions, varying object scales, and rapidly changing environments. Developing '''self-supervised''', lifelong learning systems that adapt to '''open-world''' conditions remains a pressing challenge<ref>{{Cite web |title=Frontiers {{!}} Advancing Autonomous Robots: Challenges and Innovations in Open-World Scene Understanding |url=https://www.frontiersin.org/research-topics/61975/advancing-autonomous-robots-challenges-and-innovations-in-open-world-scene-understanding |access-date=2025-08-25 |website=www.frontiersin.org}}</ref>.
'''Multi-robot coordination and decentralization'''
Scaling robot systems raises thorny issues in coordination, safety, and communication. In '''multi-agent navigation''', challenges like deadlocks, selfish behaviors, and sample inefficiencies emerge. Innovations such as dividing planning into sub-problems, combining RL with imitation learning, hybrid centralized-decentralized approaches (e.g., prioritized communication learning), attention mechanisms, and graph transformers have shown promise, but large-scale, stable, real-time coordination remains an open frontier <ref>{{Cite journal |last=Chung |first=Jaehoon |last2=Fayyad |first2=Jamil |last3=Younes |first3=Younes Al |last4=Najjaran |first4=Homayoun |date=2024-02-08 |title=Learning team-based navigation: a review of deep reinforcement learning techniques for multi-agent pathfinding |url=https://doi.org/10.1007/s10462-023-10670-6 |journal=Artificial Intelligence Review |language=en |volume=57 |issue=2 |pages=41 |doi=10.1007/s10462-023-10670-6 |issn=1573-7462}}</ref>.
'''Simulation-to-real (“reality gap”) transfer'''
[[Deep reinforcement learning]] is a powerful tool for teaching robots navigation and control, but training in simulation introduces discrepancies when deployed in reality. The '''reality gap''' (or differences between simulated and real environments) continues to impede reliable deployment, despite strategies to mitigate it<ref>{{Cite journal |last=Majid |first=Amjad Yousef |last2=van Rietbergen |first2=Tomas |last3=Prasad |first3=R Venkatesha |date=2024-08-23 |title=Challenging Conventions Towards Reliable Robot Navigation Using Deep Reinforcement Learning |url=https://scifiniti.com/3006-4163/1/2024.0003 |journal=Computing&amp;AI Connect |language=en |volume=1 |issue=1 |pages=1 |doi=10.69709/CAIC.2024.194188 |issn=3104-4719 |archive-url=http://web.archive.org/web/20250709154931/https://scifiniti.com/3006-4163/1/2024.0003 |archive-date=2025-07-09}}</ref><ref>{{Cite journal |last=Wijayathunga |first=Liyana |last2=Rassau |first2=Alexander |last3=Chai |first3=Douglas |date=2023-08-31 |title=Challenges and Solutions for Autonomous Ground Robot Scene Understanding and Navigation in Unstructured Outdoor Environments: A Review |url=https://www.mdpi.com/2076-3417/13/17/9877 |journal=Applied Sciences |language=en |volume=13 |issue=17 |pages=9877 |doi=10.3390/app13179877 |issn=2076-3417 |archive-url=http://web.archive.org/web/20241115184836/https://www.mdpi.com/2076-3417/13/17/9877 |archive-date=2024-11-15}}</ref>.
'''Hardware and bio''' '''hybrid constraints'''
Physical limitations of batteries, motors, sensors, and actuators constrain robot autonomy, endurance, and adaptability, especially for humanoid or soft-bio hybrid robots. While [[Biohybrid system]] (e.g., using living muscle tissue) hint at leveraging biological energy and actuation, they introduce radically new challenges in materials, integration, and control<ref>{{Cite web |last=Dery |first=Mikaela |date=2018-02-16 |title=10 big robotics challenges that need to be solved in the next 10 years |url=https://createdigital.org.au/robotics-challenges-next-10-years/ |access-date=2025-08-25 |website=create digital |language=en-AU}}</ref><ref>{{Cite web |title=Client Challenge |url=https://www.ft.com/content/84414ad5-6157-4f36-a27a-1366868a25ca |access-date=2025-08-25 |website=www.ft.com}}</ref>.
'''Ethics, liability, and societal integration'''
As robots become more autonomous, especially in public or collaborative roles, ethical and legal issues grow. Who is responsible when an autonomous system causes harm? Regulatory frameworks are still evolving to address liability, transparency, bias, and safety in systems like [[Self-driving car]] or socially interactive robots<ref>{{Cite news |last=Herold |first=Eve |title=How Smart Should Robots Be? |url=https://time.com/6553218/robots-ai-autonomous-essay/ |archive-url=http://web.archive.org/web/20250724094140/https://time.com/6553218/robots-ai-autonomous-essay/ |archive-date=2025-07-24 |access-date=2025-08-25 |work=TIME |language=en}}</ref>.
'''Embodied AI and industrial adoption'''
While AI algorithms have made strides, embedding them into robots (embodied AI) for real-world use remains slow-moving. Hardware constraints, economic viability, and infrastructure limitations limit widespread adoption. For instance, humanoid robots like “[[Pepper (robot)]]” failed to achieve ubiquity due to fundamental cost and complexity issues<ref>{{Cite web |title=Client Challenge |url=https://www.ft.com/content/84414ad5-6157-4f36-a27a-1366868a25ca |access-date=2025-08-25 |website=www.ft.com}}</ref><ref>{{Cite news |last=Hawkins |first=Amy |date=2025-04-21 |title=Humanoid workers and surveillance buggies: ‘embodied AI’ is reshaping daily life in China |url=https://www.theguardian.com/world/2025/apr/21/humanoid-workers-and-surveillance-buggies-embodied-ai-is-reshaping-daily-life-in-china |access-date=2025-08-25 |work=The Guardian |language=en-GB |issn=0261-3077}}</ref>.
==Societal impact and issues==
| |||