Search

Advances in information technology have led to the proliferation of data in the fields of finance, energy, and economics. Unforeseen elements can cause data to be contaminated by noise and outliers. In this study, a robust online support vector regression algorithm based on a non-convex asymmetric loss function is developed to handle the regression of noisy dynamic data streams. Inspired by pinball loss, a truncated ε -insensitive pinball loss (TIPL) is proposed to solve the problems caused by heavy noise and outliers. A TIPL-based online support vector regression algorithm (TIPOSVR) is constructed under the regularization framework, and the online gradient descent algorithm is implemented to execute it. Experiments are performed using synthetic datasets, UCI datasets, and real datasets. The results of the investigation show that in the majority of cases, the proposed algorithm is comparable, or even superior, to the comparison algorithms in terms of accuracy and robustness on datasets with different types of noise. [ABSTRACT FROM AUTHOR]

Vehicle parameters are essential for dynamic analysis and control systems. One problem of the current estimation algorithm for vehicles' parameters is that: real-time estimation methods only identify parts of vehicle parameters, whereas other parameters such as suspension damping coefficients and suspension and tire stiffnesses are assumed to be known in advance by means of an inertial parameter measurement device (IPMD). In this study, a fusion algorithm is proposed for identifying comprehensive vehicle parameters without the help of an IPMD, and vehicle parameters are divided into time-independent parameters (TIPs) and time-dependent parameters (TDPs) based on whether they change over time. TIPs are identified by a hybrid-mass state-variable (HMSV). A dual unscented Kalman filter (DUKF) is applied to update both TDPs and online states. The experiment is conducted on a real two-axle vehicle and the test data are used to estimate both TIPs and TDPs to validate the accuracy of the proposed algorithm. Numerical simulations are performed to further investigate the algorithm's performance in terms of sprung mass variation, model error because of linearization and various road conditions. The results from both the experiment and simulation show that the proposed algorithm can estimate TIPs as well as update TDPs and online states with high accuracy and quick convergence, and no requirement of road information. [ABSTRACT FROM AUTHOR]