Abstract:The Kalman filter (KF) is a state estimation algorithm that optimally combines system knowledge and measurements to minimize the mean squared error of the estimated states. While KF was initially designed for linear systems, numerous extensions of it, such as extended Kalman filter (EKF), unscented Kalman filter (UKF), cubature Kalman filter (CKF), etc., have been proposed for nonlinear systems. Although different types of nonlinear KFs have different pros and cons, they all use the same framework of linear KF, which, according to what we found in this paper, tends to give overconfident and less accurate state estimations when the measurement functions are nonlinear. Therefore, in this study, we designed a new framework for nonlinear KFs and showed theoretically and empirically that the new framework estimates the states and covariance matrix more accurately than the old one. The new framework was tested on four different nonlinear KFs and five different tasks, showcasing its ability to reduce the estimation errors by several orders of magnitude in low-measurement-noise conditions, with only about a 10 to 90% increase in computational time. All types of nonlinear KFs can benefit from the new framework, and the benefit will increase as the sensors become more and more accurate in the future. As an example, EKF, the simplest nonlinear KF that was previously believed to work poorly for strongly nonlinear systems, can now provide fast and fairly accurate state estimations with the help of the new framework. The codes are available at <a class="link-external link-https" href="https://github.com/Shida-Jiang/A-new-framework-for-nonlinear-Kalman-filters" rel="external noopener nofollow">this https URL</a>.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is that the existing Nonlinear Kalman Filters (NLKF) have the problems of over - confident and inaccurate state estimation when dealing with nonlinear measurement functions. Specifically:
1. **Limitations of the existing framework**:
- The existing NLKF frameworks (such as the Extended Kalman Filter (EKF), the Unscented Kalman Filter (UKF), etc.) can handle nonlinear systems, but they are all based on the framework of the linear Kalman filter.
- When the measurement function is nonlinear, these frameworks tend to underestimate the covariance matrix after the state update, resulting in excessive confidence and inaccurate state estimation.
2. **Specific manifestations of the problem**:
- Under the condition of low measurement noise, the actual estimation error may increase as the measurement noise decreases, which is counter - intuitive.
- The root cause of this phenomenon is that the traditional NLKF framework assumes that the Kalman gain is optimal, but in fact, this assumption is not always valid under nonlinear measurement functions.
3. **The solution proposed in the paper**:
- The author proposes a new framework to recalibrate the approximation of the nonlinear measurement function after the state update, so as to estimate the covariance matrix more accurately.
- By introducing the "recalibration" step, the new framework enables the covariance matrix to more truly reflect the influence of the Kalman gain on the state covariance and avoid excessive confidence.
4. **Verification and results**:
- The paper verifies the effectiveness of the new framework through five different application scenarios (such as 3D target tracking, terrain - referenced navigation, etc.).
- The results show that the new framework significantly improves the performance of all types of nonlinear Kalman filters, especially under the condition of low measurement noise, and the estimation error is reduced by several orders of magnitude.
In summary, this paper aims to solve the problems of over - confidence and inaccuracy of nonlinear Kalman filters when dealing with nonlinear measurement functions, and improve their performance by proposing a new framework.