There are too many methods for identification and estimation of the parameters of nonlinear static systems. But most of presented methods are with complicated theory and numerical calculations. Recently, using evolutionary optimization techniques have been noticed in nonlinear systems identification. Among these optimization techniques, we can mention Particle Swarm Optimization (PSO) method. This method unlike iterative parameter-estimation methods doesn’t require initial values of unknown parameters and unlike gradient-based methods doesn’t trap in local minimums. In recent years, the PSO method has attracted researchers attention because of employed simple operators and high rate of convergence. In order to evaluation the performance of proposed method for estimation of parameters of nonlinear static models, we consider two new cases including interpretation of gravity data in order to estimating the depth and shape factor of subsurface anomalies and also estimation 2-D robot-to-robot relative pose in multi-robot systems. Estimation of the depth and shape factor of subsurface anomalies is one of the most important tasks in geophysics sciences, so it is very important to obtain optimal methods for assignment of the depth and shape factor from gravity data. In the second case, the purpose is to calculate the relative pose of collaborator robots from robot-to-robot distance measurements and displacement estimates in multi-robot systems. Robotic researchers believe that estimating the relative pose of collaborator robots is a prerequisite for efficiently coordinating the motions of teams of robots and also solving cooperative localization problems. In order to employ PSO method in the mentioned cases, the problem of estimation of the unknown variables is addressed through multi-dimensional optimization techniques and we consider upper and lower boundaries for unknown variables of nonlinear static models based on available empiric information about practical cases. Then we distribute PSO particles such as candidate answers in predefined n-dimensional bounded space and obtain the optimized values of unknown parameters with high accuracy. Finally for evaluating the proposed method precisely, we examine obtaining results compared to Nonlinear Least Squares (NLS) method. Obtained results showed that the PSO method is an efficient method for estimating the unknown parameters of the nonlinear models and is less sensitive to the high levels of the noisy data Keyword : Particle Swarm Optimization (PSO), nonlinear static models, Relative robot-to-robot pose estimation, estimation of the depth and shape factor of subsurface anomalies
