Zero-speed detection method for individually autonomous navigation system based on hidden Markov model (HMM)

The invention discloses a zero-speed detection method for an individually autonomous navigation system based on a hidden Markov model (HMM). The method comprises the following steps: acquiring the output information of a Y-axis gyroscope in a micro inertial measurement unit of the individually autonomous navigation system; preliminarily sectioning the output information of the Y-axis gyroscope according to a function, and performing range division on the sectioned output values of the Y-axis gyroscope; sectioning the output values of the Y-axis gyroscope during the one-step movement of a user, naming the sections, and taking the sections as the state quantities of an HMM process; taking an output range as the output of the HMM to convert a zero-speed detection problem into an HMM decoding problem; acquiring a state transfer matrix A and a matrix B representative of a conversion relationship between an output process and a state process in a one-step movement process of the user; calculating the state quantity Qk of the HMM through Viterbi; and judging a zero-speed interval in the movement process of the user of the individually autonomous navigation system through an inequality. According to the zero-speed detection method, the problems are mathematized and modeled, and the detection accuracy is improved.