A Model-Based Recurrent Neural Network With Randomness for Efficient Control With Applications
A Model-Based Recurrent Neural Network With Randomness for Efficient Control With Applications
复制标题
DOI：
10.1109/tii.2018.2869588
复制
发表时间：
2019-04
影响因子：
12.3
通讯作者：
Yangming Li;Shuai Li;B. Hannaford
中科院分区：
文献类型：
--
作者：
Yangming Li;Shuai Li;B. Hannaford
近年来，冗余度机器人的递归神经网络（RNN）控制方案得到了广泛的研究。这些控制方案证明了上级的计算效率、控制精度和控制鲁棒性。但是，缺乏完整的规划。本文解释了为什么RNN控制方案会遇到这个问题。在此基础上，提出了一种新的随机RNN控制方案，该方案1）将随机性引入RNN以解决规划完备性问题; 2）通过新的优化目标提高控制精度; 3）通过探索学习提高规划效率。理论分析证明了该方法的全局稳定性、规划完备性和计算复杂性。软件仿真结果表明，改进的鲁棒性对噪声，规划的完整性和改进的规划效率的基准RNN控制方案所提出的方法。真实世界的实验来证明所提出的方法的应用。
Recently, recurrent neural network (RNN) control schemes for redundant manipulators have been extensively studied. These control schemes demonstrate superior computational efficiency, control precision, and control robustness. However, they lack planning completeness. This paper explains why RNN control schemes suffer from the problem. Based on the analysis, this work presents a new random RNN control scheme, which 1) introduces randomness into RNN to address the planning completeness problem, 2) improves control precision with a new optimization target, and 3) improves planning efficiency through learning from exploration. Theoretical analyses are used to prove the global stability, the planning completeness, and the computational complexity of the proposed method. Software simulation is provided to demonstrate the improved robustness against noise, the planning completeness and the improved planning efficiency of the proposed method over benchmark RNN control schemes. Real-world experiments are presented to demonstrate the application of the proposed method.