Intelligent control for accurate fast response and minimum energy of motion for industrial robotic manipulator
参考中译:智能控制，实现工业机器人机械手的准确快速响应和最小运动能量


          

刊名：International Journal of Automation and Control
作者：Shaar, Areej（Philadelphia Univ, Mechatron Engn Dept, Amman 19392, Jordan）
Ghaeb, Jasim A.Shaar, Areej（Philadelphia Univ, Mechatron Engn Dept, Amman 19392, JordanPhiladelphia Univ, Mechatron Engn Dept, Amman 19392, Jordan）
Ghaeb, Jasim A.（Philadelphia Univ, Mechatron Engn Dept, Amman 19392, Jordan）
刊号：738LD015
ISSN：1740-7516
出版年：2025
年卷期：2025, vol.19, no.3
页码：370-394
总页数：25
分类号：TP3
关键词：robot manipulators；；optimisation；；LQR；；linear quadratic regulator；；DOF；；degrees-of-freedom；；energy consumption；；neural network
参考中译：机器人机械手;;优化;;LQR;;线性二次调节器;;自由度;;能源消耗;;神经网络
语种：eng
文摘：This work proposes an approach to optimise the performance of a six degrees-of-freedom (6-DOF) robotic manipulator. The focus is on achieving a balance between three key objectives: rapid response speed, minimal positioning error, and reduced energy consumption during movement. The methodology employs a two-phase approach. First, a kinematic model is established using the Denavit-Hartenberg convention. Subsequently, a grey wolf optimiser (GWO) identifies optimal joint configurations for diverse target locations within the workspace. These optimal configurations serve as training data for a forward neural network (FNN) model, enabling it to predict optimal joint angles for future tasks. The proposed method demonstrates exceptional capability in precisely positioning the manipulator at desired locations within a short timeframe (0.01 sec average) while maintaining high accuracy (0.0056 mean square error (MSE) average) and achieving significant energy savings (70% average reduction). This approach presents a promising solution for enhancing the overall performance of 6-DOF robotic manipulators.
参考中译：这项工作提出了一种优化六自由度（6-DOF）机器人机械手性能的方法。重点是实现三个关键目标之间的平衡：快速响应速度、最小定位误差以及减少运动期间的能源消耗。该方法采用两阶段方法。首先，使用德纳维特-哈滕伯格惯例建立运动学模型。随后，灰狼优化器（GWO 0）识别工作空间内不同目标位置的最佳关节配置。这些最佳配置充当前向神经网络（FNN）模型的训练数据，使其能够预测未来任务的最佳关节角度。所提出的方法表现出了在短时间内（平均0.01秒）将机械手精确定位在所需位置的卓越能力，同时保持高准确性（平均0.0056均方误差（SSE）并实现显着节能（平均减少70%）。这种方法为提高6自由度机器人机械手的整体性能提供了一种有希望的解决方案。



国家科技图书文献中心  全球文献资源网  京ICP备05055788号-26  京公网安备11010202008970号  机械工业信息研究院 2018-2025