Li Yuling, Wang Liping, Liu Kun, He Wei, Yin Yixin, Johansson Rolf
IEEE Trans Neural Netw Learn Syst. 2023 Aug;34(8):4841-4855. doi: 10.1109/TNNLS.2021.3125004. Epub 2023 Aug 4.
This article addresses the distributed cooperative control design for a class of sampled-data teleoperation systems with multiple slave mobile manipulators grasping an object in the presence of communication bandwidth limitation and time delays. Discrete-time information transmission with time-varying delays is assumed, and the Round-Robin (RR) scheduling protocol is used to regulate the data transmission from the multiple slaves to the master. The control task is to guarantee the task-space position synchronization between the master and the grasped object with the mobile bases in a fixed formation. A fully distributed control strategy including neural-network-based task-space synchronization controllers and neural-network-based null-space formation controllers is proposed, where the radial basis function (RBF) neural networks with adaptive estimation of approximation errors are used to compensate the dynamical uncertainties. The stability and the synchronization/formation features of the single-master-multiple-slaves (SMMS) teleoperation system are analyzed, and the relationship among the control parameters, the upper bound of the time delays, and the maximum allowable sampling interval is established. Experiments are implemented to validate the effectiveness of the proposed control algorithm.
本文针对一类存在通信带宽限制和时延的、具有多个从移动机械手抓取物体的采样数据遥操作系统，研究其分布式协同控制设计。假设采用具有时变时延的离散时间信息传输，并使用循环（RR）调度协议来调节从多个从设备到主设备的数据传输。控制任务是保证主设备与抓取物体之间在任务空间位置同步，且移动基座保持固定编队。提出了一种完全分布式控制策略，包括基于神经网络的任务空间同步控制器和基于神经网络的零空间编队控制器，其中采用具有近似误差自适应估计的径向基函数（RBF）神经网络来补偿动态不确定性。分析了单主多从（SMMS）遥操作系统的稳定性以及同步/编队特性，并建立了控制参数、时延上限和最大允许采样间隔之间的关系。通过实验验证了所提控制算法的有效性。
