Robust Online Model Adaptation by Extended Kalman Filter with Exponential Moving Average and Dynamic Multi-Epoch Strategy

EDU Robotics Institute, Carnegie Mellon University, Pittsburgh, P A 15213, USA Abstract High fidelity behavior prediction of intelligent agents is critical in many applications. However, the prediction model trained on the training set may not generalize to the testing set due to domain shift and time variance. The challenge motivates the adoption of online adaptation algorithms to update prediction models in real-time to improve the prediction performance. Inspired by Extended Kalman Filter (EKF), this paper introduces a series of online adaptation methods, which are applicable to neural network-based models. A base adaptation algorithm Modified EKF with forgetting factor (MEKF λ) is introduced first, followed by exponential moving average filtering techniques. Then this paper introduces a dynamic multi-epoch update strategy to effectively utilize samples received in real time. With all these extensions, we propose a robust online adaptation algorithm: MEKF with Exponential Moving Average and Dynamic Multi-Epoch strategy (MEKF EMA-DME). The proposed algorithm outperforms existing methods as demonstrated in experiments. Keywords: Online adaptation, extended Kalman filter, exponential moving average, optimization 1. Introduction Supervised learning has been widely used to obtain models to predict the behaviors of intelligent agents Rudenko et al. (2019). Behavior prediction is a sub-topic of time series prediction Weigend (2018), which includes but is not limited to vehicle trajectory prediction during autonomous driving Lef evre et al. (2014) and human-motion prediction during human-robot collaboration Cheng et al. (2019).