Simultaneous hand/wrist motion recognition and continuous grasp force estimation based on nonlinear spectral sEMG features for transradial amputees

•A framework is proposed for simultaneous motion classification and continuous grasp force estimation to enhance the natural and robust control of multiple degrees of freedom (MDOF) myoelectric prostheses.•A new set of nonlinear mel-frequency spectrum features of EMG signals was designed to achieve robust EMG pattern recognition under varying contraction forces and accurate estimation of continuous grasp force.•The proposed method was tested on both able bodied subjects and amputees who are the end-users of prosthesis.•The performance of the proposed feature set was systematically compared to that of the state-of-art EMG feature sets.•Experimental results suggested that the proposed method would be beneficial for stable and fine manipulation of multifunctional prosthetic hand.