Online Residual Learning from Offline Experts for Pedestrian Tracking
This work addresses pedestrian tracking for autonomous systems, but it is incremental as it builds on existing prediction frameworks with minor enhancements.
The paper tackles the problem of online pedestrian trajectory prediction by proposing Online Residual Learning (ORL), which combines offline-trained predictions with online adaptation to reduce error, resulting in improved performance on the Stanford Drone Dataset.
In this paper, we consider the problem of predicting unknown targets from data. We propose Online Residual Learning (ORL), a method that combines online adaptation with offline-trained predictions. At a lower level, we employ multiple offline predictions generated before or at the beginning of the prediction horizon. We augment every offline prediction by learning their respective residual error concerning the true target state online, using the recursive least squares algorithm. At a higher level, we treat the augmented lower-level predictors as experts, adopting the Prediction with Expert Advice framework. We utilize an adaptive softmax weighting scheme to form an aggregate prediction and provide guarantees for ORL in terms of regret. We employ ORL to boost performance in the setting of online pedestrian trajectory prediction. Based on data from the Stanford Drone Dataset, we show that ORL can demonstrate best-of-both-worlds performance.