The invention discloses a gripping function evaluating and training 
system. The gripping function evaluating and training 
system comprises a host interaction module, a 
signal acquisition module, a 
signal storage module and a host control module, wherein the host interaction module is used for guiding a user to adjust gripping force and 
muscle contraction conditions in real time to fulfill a virtual task; the 
signal acquisition module is used for acquiring mechanical signals and electromyographic signals of the upper limbs of the user; the signal storage module is used for receiving and storing the mechanical signals and the electromyographic signals of the 
signal acquisition module, and storing characteristic parameters; the host control module is connected with the host interaction module and the signal storage module respectively, and is used for computing and 
processing the mechanical signals and the electromyographic signals transmitted by the 
signal acquisition module to obtain the characteristic parameters, controlling a virtual scene to perform corresponding game operation so as to 
train the 
motion control function through the characteristic parameters, and finally transmitting the characteristic parameters to the signal storage module. The gripping function evaluating and training 
system performs human-computer interaction by utilizing multiple parameters such as the electromyographic signals and the mechanical signals, and is used for training and evaluating the 
hand function of a 
stroke patient, so that the 
hand motion function of the patient is improved.