Abstract: In order to solve the problem of the identification accuracy of dynamic parameters of mine water filling water filling robotic arm, an improved ISSA algorithm was proposed to identify the dynamic parameters.In the process of improving ISSA algorithm, Logistic mapping function was improved to generate diverse variable populations and improve the diversity and adaptability of identified populations.The dynamic proportional adjustment function is adopted to achieve comprehensive identification of the population; Search location update strategy is improved to boost the convergence accuracy of the algorithm; The tent perturbation and Cauchy perturbation functions were introduced to enhance the adaptability to the identification of populations.Through the dynamic identification experiment of robotic arm, the excitation trajectories, joint moments, algorithm convergence diagrams and algorithm results of different joints were obtained.The results show that the improved ISSA algorithm has obvious performance in terms of accuracy and convergence speed, which provides a new solution to the problem of dynamic parameter identification.