Aiming at the problem that the error parameters of the gyroscope are difficult to identify due to the influence of the measurement environment while drilling, an improved meerkat optimization algorithm (IMOA) is proposed to compensate for the gyroscope error. Firstly, the gyroscope error model is derived and the error parameters that need to be identified are determined. The objective function is established by using the output characteristics of the accelerometer, and the constraint conditions are set according to the inner product characteristics and the relative error of the magnetic modulus value. The meerkat optimization algorithm (MOA) is adopted to solve the optimal value of the error vector. Based on the MOA algorithm, the step size is reset with the relative modulus error of the accelerometer to adaptively track the changes in the gyroscope signal caused by the drilling environment. A guided spiral global update method is designed. During the update, a guiding factor is set to judge the superiority or inferiority of the current search direction, and a per-dimensional reward mechanism is utilized to ensure that different parameters retain the optimal solution to a large extent when exchanging information, preventing falling into local optimum. Set the local migration route and redefine the deviation as the cartesian distance between the error parameter of the current individual identification and the historical optimal error parameter; Meanwhile, the per-dimensional perturbation strategy is added to retain the optimal solutions of the error parameters in each dimension. Finally, the IMOA algorithm was applied to identify the error parameters of the gyroscope. The results show that the output error of the gyroscope after compensation by the IMOA algorithm is significantly reduced, and the well slope angle error is reduced from 9.54° to 2.53°. Compared with the PSO algorithm and the MOA algorithm, it has higher recognition accuracy.