“…- Maximum likelihood identification method (Young, 2015).
- The closed-loop output error method and its variants (Gautier et al , 2013; Janot et al , 2014a, 2014b; Brunot et al , 2020): the closed-loop input error and the direct and inverse dynamic identification model algorithm.
- The direct dynamics identification model with Kalman filtering method and its improved methods (Joukov et al , 2015; Riva et al , 2017), namely, the extended Kalman filter, the unscented Kalman filter, the central difference Kalman filter, the bootstrap particle filter, square-root extended Kalman filter, square-root unscented Kalman filter and square-root central difference Kalman filter.
- Neural networks methods (Leboutet et al , 2021), such as Adaline neural network and Hopfield–Tank recurrent neural networks.
…”