Tactile Compensation for Artificial Whiskered Sensor System Under Critical Change in Morphology
Tactile Compensation for Artificial Whiskered Sensor System Under Critical Change in Morphology
复制标题
DOI：
10.1109/lra.2021.3064460
复制
发表时间：
2021-04
影响因子：
5.2
通讯作者：
Nhan Huu Nguyen;V. A. Ho
中科院分区：
文献类型：
--
作者：
Nhan Huu Nguyen;V. A. Ho
在这封信中，我们描述了一种新的设计软whiskersensor系统，补偿触觉缺陷的传感器形态变化的事件后，被打破，撕裂，或修剪。传感器根部的纤维增强气室与应变计和压缩机耦合。当触须接触物体时，应变仪向中央控制系统（CCS）提供触觉信息。如果晶须在距其尖端一定长度处被修剪，为了适应这种变化，腔室中的空气压力改变其形态，引起应变仪灵敏度的变化，从而补偿晶须长度损失。应用有限元法（FEM）分析和遗传算法（GA）优化来确定满足所有性能要求的最佳设计，特别是与补偿能力（即自适应范围）相关的性能要求。进行了适应范围为5 mm和10 mm的两次试验，并通过实验验证了前者对应的适应度得分大于后者的最优设计。实验结果表明，触觉传感性能约87$\%$和72$\%$的恢复，分别显示所提出的方法的可靠性。这项研究加强了这一概念，形态控制的应用有可能带来额外的能力，在应用软机器人系统的CCS上的负担最小。
In this letter, we describe a novel design soft whiskersensor system that compensates for tactile deficiency in events of change in sensor morphology upon being broken, torn, or trimmed. A fiber-reinforced air chamber at the sensor root is coupled with a strain gauge, and compressor. The strain gauge provides tactile information to the central control system (CCS) when the whisker contacts an object. If the whisker is trimmed at a certain length from its tip, to adapt for this change, air pressure in the chamber alters its morphology inducing a change in the sensitivity of the strain gauge that compensates for whisker length loss. Finite element method (FEM) analysis and genetic algorithm (GA) optimization were applied to determine optimal designs satisfying all performance requirements especially related to compensation ability, namely adaptive range. Two trials of adaptive ranges 5 mm and 10 mm were conducted, and optimal designs, where the corresponding fitness score of the former is bigger than the latter, were experimentally validated. Experimental results demonstrated tactile sensing performance of approximately 87$\%$ and 72$\%$ recovery, respectively, showing reliability of the proposed method. This study reinforces the concept that application of morphological control has potential to bring added abilities in applied soft robotic systems with minimum burden on the CCS.