Scholarlink

<p>Industrial manipulators have evolved from rigid, pre-programmed devices into adaptive and intelligent systems shaped by advances in artificial intelligence (AI), soft robotics, advanced sensing, and collaborative technologies. This review presents a multidimensional taxonomy of manipulators that integrates kinematic structures, actuation principles, control strategies, and human–robot interaction capabilities. Particular emphasis is placed on advanced control frameworks, including machine learning-based adaptive strategies and hybrid optimization methods for localization and geometric error reduction. Perception-driven approaches, such as deep convolutional neural networks for indoor scene recognition, are enabling more autonomous operation in unstructured environments. The rise of soft robotics and compliant actuation enhances adaptability and safety in collaborative contexts. We also address persistent challenges, including computational complexity, energy efficiency, economic scalability, and standardization gaps in safety frameworks. By synthesizing current literature and global industrial reports, this work highlights both incremental progress and disruptive opportunities, outlining future research directions such as quantum computing integration, advanced sensing for Industry 5.0, and AI-driven self-learning control.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Advances in intelligent industrial manipulators for smart manufacturing and standardized automation technologies

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月20日

查看全文
  • Ahmed Sameh,
  • Mohamed Fanni,
  • Maher Rashad

摘要

Industrial manipulators have evolved from rigid, pre-programmed devices into adaptive and intelligent systems shaped by advances in artificial intelligence (AI), soft robotics, advanced sensing, and collaborative technologies. This review presents a multidimensional taxonomy of manipulators that integrates kinematic structures, actuation principles, control strategies, and human–robot interaction capabilities. Particular emphasis is placed on advanced control frameworks, including machine learning-based adaptive strategies and hybrid optimization methods for localization and geometric error reduction. Perception-driven approaches, such as deep convolutional neural networks for indoor scene recognition, are enabling more autonomous operation in unstructured environments. The rise of soft robotics and compliant actuation enhances adaptability and safety in collaborative contexts. We also address persistent challenges, including computational complexity, energy efficiency, economic scalability, and standardization gaps in safety frameworks. By synthesizing current literature and global industrial reports, this work highlights both incremental progress and disruptive opportunities, outlining future research directions such as quantum computing integration, advanced sensing for Industry 5.0, and AI-driven self-learning control.