Adaptive grasping control of bionic arm via electromyographic signals
仿生机械臂的肌电信号自适应抓取控制

摘要

聚焦仿生机械臂的肌电信号自适应抓取控制研究。先阐述通过肌电传感器获取人体肌肉运动产生的微弱电信号，经滤波、放大等预处理步骤提高信号质量。接着介绍从预处理信号中提取时域、频域及时频域特征以识别肌肉运动意图的方法。然后说明基于识别结果设计控制算法，如采用自适应控制策略使机械臂能依据抓取物体的特性及环境变化实时调整控制参数，实现稳定抓取。研究旨在提升仿生机械臂抓取的准确性、灵活性与适应性，推动其在医疗康复、工业生产等多领域的广泛应用。

Abstract

This study focuses on the research of adaptive grasping control for bionic arms using electromyographic signals. The paper first explains how microelectromyographic signals generated by human muscle movements are acquired through electromyography sensors, followed by signal quality enhancement through preprocessing steps such as filtering and amplification. It then introduces methods for extracting time-domain, frequency-domain, and time-frequency domain features from the processed signals to identify muscle movement intentions. Subsequently, control algorithms are designed based on recognition results, including adaptive control strategies that enable real-time adjustment of control parameters according to object characteristics and environmental changes, achieving stable grasping. The research aims to enhance the accuracy, flexibility, and adaptability of bionic arm grasping, promoting its widespread application in medical rehabilitation, industrial production, and other fields.