A shape memory alloy-based biomimetic robotic hand : design, modelling and experimental evaluation
Filomena Simone
- 发表年份
- 2021
- 引用次数
- 2
- 访问权限
- 开放获取
摘要
Every year more the 400,000 people are subject to an upper limb amputation. Projections foresee that this number may double by the 2050. Infections, trauma, cancer, or complications that arise in blood vessels represent the main causes for amputations. The access to prosthetic care is worldwide extremely limited. This is mainly due to the high cost both of commercially available prostheses and of the rehabilitation procedure which every prostheses user has to endure. Aside from high costs, commercially available hand prostheses have faced high rejection rates, mainly due to the their heavy weight, noisy operation and also to the unnatural feel of the fingers. To overcome these limitations, new materials, such as Shape Memory Alloys (SMAs), have been considered as potential candidate actuators for these kind of devices. In order to provide a contribution in the development of performant and easily affordable hand prostheses, the development of a novel and cost-effective five-fingered hand prototype actuated by Shape Memory Alloy (SMA) wires is presented in this work. The dissertation starts with the description of a first generation of a SMA actuated finger. Structure assemblage and performances in term of force, motion and reactiveness are investigated to highlight advantages and disadvantages of the prototype. In order to improve the achievable performances, a second generation of SMA actuated finger having soft features is introduced. Its structure, a five-fingered hand prosthesis having intrinsically elastic fingers, capable to grasp several types of objects with a considerable force, and an entirely 3D printed structure is then presented. Comparing this prototype with the most important prostheses developed so far, relevant advantages especially in term of noiseless actuation, cost, weight, responsiveness and force can be highlighted. A finite element based framework is then developed, to enable additional structure optimization and further improve the SMA finger performances. On the same time, a concentrated parameters physics-based model is formulated to allow, in the future, an easier control of the device, characterized by strong nonlinearities mainly due to the Shape Memory alloy hysteretic behavior.
关键词
相关论文
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002