This paper studies an adaptive approach for probabilistic wind power forecasting (WPF) including offline and online learning procedures. In the offline learning stage, a base forecast model is trained via inner and outer loop updates of meta-learning, which endows the base forecast model with excellent adaptability to different forecast tasks, i.e., probabilistic WPF with different lead times or locations. In the online learning stage, the base forecast model is applied to online forecasting combined with incremental learning techniques. On this basis, the online forecast takes full advantage of recent information and the adaptability of the base forecast model. Two applications are developed based on our proposed approach concerning forecasting with different lead times (temporal adaptation) and forecasting for newly established wind farms (spatial adaptation), respectively. Numerical tests were conducted on real-world wind power data sets. Simulation results validate the advantages in adaptivity of the proposed methods compared with existing alternatives.

Noticing and Perception is a critical element of an autonomous system, specifically when deployed in very dynamic environments where it is needed to respond to unforeseen circumstances. This restriction does not aid designers and third party evaluators to answer a vital concern: is the performance of an assumption subsystem sufficient for the choice making subsystem to make robust, secure choices? At the exact same time, we show how to analyze the effect of various kinds of sensing and assumption mistakes on the habits of the autonomous system. Motion prediction of vehicles is critical yet challenging because of the uncertainties in complex environments and the limited visibility triggered by occlusions and limited sensing unit arrays. Unlike the existing trajectory forecast task for seen vehicles, we target at predicting a tenancy map that suggests the earliest time when each location can be inhabited by either seen and hidden vehicles.