CHARTING THE RESEARCH TERRAIN OF INTENTION RECOGNITION THROUGH CITESPACE ANALYSIS

Abstract

Intention recognition (IR) plays a vital role in various domains like human-robot interaction, human-computer interaction, and human-vehicle interaction. It is crucial for enhancing the efficiency of human-robot collaboration, as seen in the application of rehabilitation robots. Researchers leverage Electromyography (EMG) to capture lower limb neural information, reflecting human intentions. Hidden Markov Models (HMM) have been employed to predict these intentions. Machine learning algorithms, including attention-based Long Short-Term Memory (LSTM) Networks, Convolutional Neural Networks, and Herman Neural Networks, are used to classify intentions accurately. Recent advancements in sensor technology have improved human motion recognition and the monitoring of muscular activity. This has led to a substantial growth in the literature related to IR. To facilitate a comprehensive understanding of this research landscape, tools like CiteSpace are employed to create knowledge maps, highlighting research trends and frequently used methods.