To accurately identify human lower limb gait movements, a method to identify the swing phase and support phase of lower limb gait is designed. The human lower limb angle information and plantar pressure information are collected through four posture sensors and plantar pressure insoles, and the data information is normalized and scaled to extract features; the sensor information is fuzzified by using the fuzzy principle to classify the gait of both legs into four cases; the support vector machine (SVM) with different kernel functions is used to recognize the lower limb angle information and plantar pressure information using MATLAB; the same gait of a person walking in a straight line at different gait rates and the gait of a person of different height and leg length walking in a straight line at a rate of 0.6 m/s were experimented. The results show that the algorithm is effective and applicable, and the average recognition accuracy is above 90% in all cases.