ABSTRACT
This chapter presents a multiclass fuzzy relevance vector machine (FRVM) learning mechanism and evaluates its performance to classify multiple hand motions using surface electromyographic (SEMG) signals. The relevance vector machine (RVM) is a sparse Bayesian kernel method, which avoids some limitations of the support vector machine (SVM). However, the RVM still suffers from the difficulty of possible unclassifiable regions in multiclass problems. We propose two fuzzy membership function–based FRVM algorithms to solve such problems, based on experiments conducted on seven subjects with six hand motions. Two feature sets, namely, time domain (TD) and wavelet transform (WT) features, are extracted from the recorded electromyographic (EMG) signals. Fuzzy support vector machine (FSVM) analysis was also conducted for comparison. For both TD and WT features, FRVM demonstrates less sensitivity to membership functions, while FSVM provides quite-different classification accuracies when using various membership functions. FRVM yielded comparable classification accuracy with dramatically fewer support vectors in comparison with FSVM. Furthermore, the processing delay of FRVM was much less than that of FSVM. The results indicate that an FRVM classifier can achieve comparable generalization capability as FSVM with significant sparsity in multichannel EMG classification, which is more suitable for EMG-based real-time control applications.
