The dexterous upper limb serves as the most important tool for astronauts to implement in-orbit experiments and operations.This study developed a simulated weightlessness experiment and invented new measuring equipment to quantitatively evaluate the muscle ability of the upper limb.Isometric maximum voluntary contractions(MVCs) and surface electromyography(sEMG) signals of right-handed pushing at the three positions were measured for eleven subjects.In order to enhance the comprehensiveness and accuracy of muscle force assessment,the study focused on signal processing techniques.We applied a combination method,which consists of time-,frequency-,and bi-frequency-domain analyses.Time-and frequency-domain analyses estimated the root mean square(RMS) and median frequency(MDF) of sEMG signals,respectively.Higher order spectra(HOS) of bi-frequency domain evaluated the maximum bispectrum amplitude(Bmax),Gaussianity level(Sg) and linearity level(Sl) of sEMG signals.Results showed that B max,S l,and RMS values all increased as force increased.MDF and S g values both declined as force increased.The research demonstrated that the combination method is superior to the conventional time-and frequency-domain analyses.The method not only described sEMG signal amplitude and power spectrum,but also deeper characterized phase coupling information and non-Gaussianity and non-linearity levels of sEMG,compared to two conventional analyses.The finding from the study can aid ergonomist to estimate astronaut muscle performance,so as to optimize in-orbit operation efficacy and minimize musculoskeletal injuries. The dexterous upper limb serves as the most important tool for astronauts to implement in-orbit experiments and operations. This study developed a simulated weightlessness experiment and invented new measuring equipment to quantitatively evaluate the muscle ability of the upper limb. Isometric maximum voluntary contractions (MVCs ) and surface electromyography (sEMG) signals of right-handed pushing at the three positions were measured for eleven subjects. In order to enhance the comprehensiveness and accuracy of muscle force assessment, the study focused on signal processing techniques. We applied a combination method, which consists of time-, frequency-, and bi-frequency-domain analyzes. Time-and frequency-domain analyzes estimated the root mean square (RMS) and median frequency (MDF) of sEMG signals, respectively .Higher order spectra (HOS) of bi-frequency domain as the maximum bispectrum amplitude (Bmax), Gaussianity level (Sg) and linearity level (S1) of sEMG signals. Results showed that B max, S l, and RMS values ​​all increased as force increased. .MDF and S g values ​​both declined as force increased. The research demonstrated that the combination method is superior to the conventional time-and frequency-domain analyzes. The method not only described sEMG signal amplitude and power spectrum, but also further characterized phase coupling information and non-Gaussianity and non-linearity levels of sEMG, compared to two conventional analyzes. the finding from the study can aid ergonomist to estimate astronaut muscle performance, so as to optimize in-orbit operation efficacy and minimize musculoskeletal injuries.