基于捕蝇草膨压-快速屈曲运动机制的仿生柔性叶片驱动器研究

The precise control and monitoring technology of pests are of great significance to the protection of the forest. How to combine robot technology to achieve automatic trapping is a difficult point. The flytrap blade has a fast moving speed and a reversible action. The flexible capture robot developed by its mechanism has potential application value in the field of forestry pest control. The thigmonastic movements of some plant leaves have the advantages of good flexibility, high velocity, reversibility, and large amplitude. The software robot developed based on the mechanism can realize flexible grasping and other actions, which is conducive to the realization of human, machine and environment integration. Firstly, the anatomical and ultrastructural analysis of plant leaves is carried out. The tissue distribution, microchannel structure and asymmetry characteristics are studied. The micro-fluid motion driven by the chemical potential at the microscopic level and the quantitative kinematics of the leaves at the macroscopic level are studied. Secondly, the turgor -snap buckling movement mechanism of mechanically instable leaves is studied, including the reversible expansion/contraction law of cells and tissues driven by water, the influence of structural features on the triggering threshold of rapid motion, the amplitude velocity variation of rapid switching between concave and convex curvature. Finally, the magneto-rheological fluid is selected as the transmission medium, and the bionic leaf driver is prepared by 3D printing technology, which contains the micro-flow channel structure and the initial elastic curvature. Through the pressure control and magnetic field control on the bionic leaf, the process of rapid curvature change and stiffness maintenance is realized. This study can solve the problem that the single hydraulic drive action is not fast enough, the contradiction between flexible gripping and rigidity retention. It also provide theoretical basis and technical support for the rapid capture research of pests.

害虫的精准防治和监测技术对森林保护意义重大，如何结合机器人技术实现自动化诱捕是一个难点。捕蝇草叶片运动速度快，动作可逆，借鉴其机制研制的柔性捕捉机器人，具备林业害虫防治领域的潜在应用价值。首先对捕蝇草叶片进行解剖和超微结构分析，研究其组织分布、微流道结构及不对称特征，研究微观层面上的微流体运动及宏观层面上叶片的定量运动学。其次研究机械失稳叶片的膨压-快速屈曲运动机制，包括水驱动下细胞及组织可逆膨胀/收缩规律，结构特征对屈曲触发阈值的影响规律，凹凸曲率快速切换的幅度及速度变化规律等。最后选择磁流体为传动介质，利用3D打印技术制备仿生叶片驱动器，具备微流道网格结构及初始弹性曲率。通过压力控制和磁场控制实现仿生叶片的磁流体压力驱动、快速曲率变化、磁流体固化保持的动作全过程。本研究可解决单一液压驱动动作不够迅速，柔性抓取和刚性保持存在矛盾的问题，为害虫的快速捕捉研究提供理论基础和技术支持。

害虫的精准防治和监测技术对森林保护意义重大，如何结合机器人技术实现自动化诱捕是一个难点。捕蝇草叶片运动速度快，动作可逆，借鉴其机制研制的柔性捕捉机器人，具备林业害虫防治领域的潜在应用价值。本项目首先进行了捕蝇草叶片微观结构、微流道的流场特征研究，得到动作规律。通过实验分析捕蝇草叶片闭合过程的应变分布，结果表明叶片的快速闭合运动主要由垂直于中脉上的变形引起。采用水弹性曲率模型对叶片闭合运动的过程进行建模，基于多孔介质模型进行细胞水分传输仿真分析。基于捕蝇草的三维叶片模型，进行仿生叶片结构设计，建立数学模型，制定仿生捕蝇草叶片的腔室设计规则。以硅橡胶作为材料，采用3D打印模具及硅胶复合浇筑的方式制作仿生叶片驱动器。完成SMA驱动和气动驱动的仿生叶片样机制作，进行弯曲性能实验。SMA丝驱动仿生捕蝇草叶片张开角度58°，所需响应时间为7s，满足技术要求。气动仿生捕蝇草样机从初始状态到闭合状态，从闭合状态到初始状态的总时间需4s左右，能够对抓取物实现包裹或者半包裹式自适应抓取。开展基于磁流变液驱动的变刚度仿生捕蝇草的设计与实验研究，分析磁流变液的相关特性，设计制作了基于磁流变液驱动的仿生捕蝇草样机，在亥姆霍兹线圈电流从0A到30A的条件下，闭合力提高了104.4%。在磁场作用下，磁流变液驱动的仿生捕蝇草包络抓取力提高了48.19%，指尖抓取力提高了17.11%。根据果蔬采摘任务的特性，探究了仿生柔性叶片驱动器在无损采摘领域的应用，提出了基于草莓曲线特征的软体抓手，进行有限元仿真分析，优化软体抓手的气体通道结构，成功抓取的概率达到90%，表皮破损率为2%，有效解决了抓取不牢和容易损伤目标物表面的问题。最后在仿生柔性叶片驱动器的研究基础上，受藤蔓缠绕运动启发，设计了一种仿生缠绕软体执行器，能够实现弯曲和缠绕运动。建立仿生藤蔓软体执行器运动过程的数学模型，通过实验得到执行器缠绕力能够达到0.77N，弯曲状态阻塞力可达到1.68N。