abstract
- Estimation theory is an important field in mechanical and electrical engineering, and is comprised of strategies that are used to predict, estimate, or smooth out important system state and parameters. The most popular and well-studied estimation strategy was developed over 60 years ago, and is referred to as the Kalman filter (KF). The KF yields the optimal solution in terms of estimation error for linear, well-known systems. Other variants of the KF have been developed to handle modeling uncertainties, non-Gaussian noise, and nonlinear systems and measurements. Although KF-based methods typically work well, they lack robustness to uncertainties and external disturbances - which are prevalent in signal processing and target tracking problems. The smooth variable structure filter (SVSF) was introduced in an effort to provide a more robust estimation strategy. In an effort to improve the robustness and filtering strategy further, this paper introduces an adaptive form of the SVSF based on the static multiple model strategy.