✦ LIBER ✦

Trainable command recognition for a microprocessor-controlled prosthetic arm

✍ Scribed by Christof Schärer

Publisher: Elsevier Science
Year: 1993
Weight: 170 KB
Volume: 16
Category: Article
ISSN: 0745-7138
DOI: 10.1006/jmca.1993.1029

No coin nor oath required. For personal study only.

✦ Synopsis

The complexity of controlling a prosthetic arm with several degrees of freedom is reduced by incorporating a microprocessor-based controller. Software algorithms can relieve the user from part of the learning effort. Adaptation to the device is divided into user learning and complementary machine learning. Machine learning takes place using an artificial neural network implemented on a digital microprocessor. Tests have shown the feasibility of the approach. For improved prosthesis control command signal acquisition has to be refined.

📜 SIMILAR VOLUMES

131 EMG pattern recognition by neural ne

131 EMG pattern recognition by neural networks for prosthetic fingers control: A. Hiraiwa, N. Uchida, K. Shimohara, pp 73–80

📂 Article 📅 1993 🏛 Elsevier Science 🌐 English ⚖ 116 KB

The contribution of this paper is twofold. First, it focuses on the application of a particular NARMAX (nonlinear ARMAX) model representation based on local models for adaptive decoupling. Second, in order to improve the robusmess of the adaptive control algorithm, a diagonal PI-controller is introd