Gait recognition is crucial for the human-machine integration of smart wearable devices.The traditional gait recognition scheme has slow response speed and low accuracy of comprehensive recognition of multiple gaits

resulting in low human-machine integration and poor wearing comfort.Aiming at the requirement of high human-machine integration of wearable devices

a accurate gait and phase recognition method combining surface electromyography(sEMG)sensor and inertial measurement unit(IMU)is proposed.With the three-axis acceleration sequence as the basic feature source

a multilayer perceptron(MLP)neural network model is constructed to classify and identify six gaits of walking

running

going upstairs

going downstairs

going uphill

and going downhill.The sEMG sensor is introduced synchronously to study the correlation logic between gastrocnemius muscle contraction relaxation and gait phase

and the precise judgment of gait phase is realized through the cooperative mechanism of "inertial feature-electromyography intensity-edge computing".The test results show that the gait recognition response time is less than 0.76s

and the comprehensive recognition accuracy reaches 96.57%.This method provides key technical support for the intelligent and precise human-machine integration of wearable devices

and has certain engineering application value.