On-Device Adaptive Battery Power Prediction for Electric Vehicles

• URL: http://arxiv.org/abs/2607.09400v1
• Date: Fri, 10 Jul 2026 13:24:34 GMT
• Title: On-Device Adaptive Battery Power Prediction for Electric Vehicles
• Abstract summary: We introduce a novel approach that enables on-device learning in resource-constrained EV systems to continuously adapt pretrained battery prediction models to new, unseen data.<n>Our results demonstrate significant improvements in forecasting performance across various models and time horizons.
• Score: 0.5314069314483559
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Adaptive power management in Electric Vehicles (EVs) requires accurate power prediction. Although deep learning models have emerged as highly effective for time-series forecasting in this domain, their performance is prone to degradation when exposed to data with distributions different from the training data. We introduce a novel approach that enables on-device learning in resource-constrained EV systems to continuously adapt pretrained battery prediction models to new, unseen data. We leverage existing pretrained models by transforming them into adaptable versions that retain critical hyperparameter knowledge from their initial training. We comprehensively investigate both online and offline model adaptation strategies. Our results demonstrate significant improvements in forecasting performance across various models and time horizons, achieving mean absolute error reductions of up to 7.49\% and 14.88\% with online and offline adaptation techniques, respectively. This study highlights the substantial benefit of on-device adaptation, resulting in enhanced battery power predictions than unadapted model deployments in real-world EV scenarios.

Related papers

• Amortized Low-Rank Adaptation for Model-Based Reinforcement Learning [49.3087425299849]
  World models let agents plan by predicting the consequences of their actions, but changes in the environment can make them inaccurate.<n>Existing approaches trade off computational cost against expressivity, i.e., the range of models a method can produce.<n>We present CLAW, which addresses this tradeoff by using a hypernetwork to generate adapters at test time.
  arXiv  Detail & Related papers  (2026-09-10T23:16:40Z)
• MetaDAT: Generalizable Trajectory Prediction via Meta Pre-training and Data-Adaptive Test-Time Updating [18.84791793147564]
  Existing trajectory prediction methods exhibit significant performance degradation under distribution shifts during test time.<n>We propose a meta-learning framework to directly optimize the predictor for fast and accurate online adaptation.<n>At test time, we introduce a data-adaptive model updating mechanism that dynamically adjusts the learning rates and updating frequencies based on online partial derivatives and hard sample selection.<n>Results demonstrate that our method achieves superior adaptation accuracy, surpassing state-of-the-art test-time training methods for trajectory prediction.
  arXiv  Detail & Related papers  (2026-03-10T09:34:32Z)
• IISE PG&E Energy Analytics Challenge 2025: Hourly-Binned Regression Models Beat Transformers in Load Forecasting [0.0]
  This study evaluates forecasting models ranging from classical regression techniques to advanced deep learning architectures.<n>The dataset includes two years of historical electricity load data, alongside temperature and global horizontal irradiance (GHI) across five sites.<n>Our results reveal that deep learning models, including TimeGPT, fail to consistently outperform simpler statistical and machine learning approaches.
  arXiv  Detail & Related papers  (2025-05-16T15:55:34Z)
• ReAugment: Model Zoo-Guided RL for Few-Shot Time Series Augmentation and Forecasting [74.00765474305288]
  We present a pilot study on using reinforcement learning (RL) for time series data augmentation.<n>Our method, ReAugment, tackles three critical questions: which parts of the training set should be augmented, how the augmentation should be performed, and what advantages RL brings to the process.
  arXiv  Detail & Related papers  (2024-09-10T07:34:19Z)
• Forecast-PEFT: Parameter-Efficient Fine-Tuning for Pre-trained Motion Forecasting Models [68.23649978697027]
  Forecast-PEFT is a fine-tuning strategy that freezes the majority of the model's parameters, focusing adjustments on newly introduced prompts and adapters. Our experiments show that Forecast-PEFT outperforms traditional full fine-tuning methods in motion prediction tasks. Forecast-FT further improves prediction performance, evidencing up to a 9.6% enhancement over conventional baseline methods.
  arXiv  Detail & Related papers  (2024-07-28T19:18:59Z)
• Test-Time Model Adaptation with Only Forward Passes [68.11784295706995]
  Test-time adaptation has proven effective in adapting a given trained model to unseen test samples with potential distribution shifts. We propose a test-time Forward-Optimization Adaptation (FOA) method. FOA runs on quantized 8-bit ViT, outperforms gradient-based TENT on full-precision 32-bit ViT, and achieves an up to 24-fold memory reduction on ImageNet-C.
  arXiv  Detail & Related papers  (2024-04-02T05:34:33Z)
• Secure short-term load forecasting for smart grids with transformer-based federated learning [0.0]
  Electricity load forecasting is an essential task within smart grids to assist demand and supply balance. Fine-grained load profiles can expose users' electricity consumption behaviors, which raises privacy and security concerns. This paper presents a novel transformer-based deep learning approach with federated learning for short-term electricity load prediction.
  arXiv  Detail & Related papers  (2023-10-26T15:27:55Z)
• A physics-informed and attention-based graph learning approach for regional electric vehicle charging demand prediction [7.441576351434805]
  This paper proposes a novel approach that enables the integration of graph and temporal attention mechanisms for feature extraction. Evaluation results on a dataset of 18,013 EV charging piles in Shenzhen, China show that the proposed approach, named PAG, can achieve state-of-the-art forecasting performance.
  arXiv  Detail & Related papers  (2023-09-11T06:31:45Z)
• An Adaptive Deep Learning Framework for Day-ahead Forecasting of Photovoltaic Power Generation [0.8702432681310401]
  This paper proposes an adaptive LSTM (AD-LSTM) model, which is a DL framework that can not only acquire general knowledge from historical data, but also dynamically learn specific knowledge from newly-arrived data. The developed AD-LSTM model demonstrates greater forecasting capability than the offline LSTM model, particularly in the presence of concept drift.
  arXiv  Detail & Related papers  (2021-09-28T02:39:56Z)
• Back2Future: Leveraging Backfill Dynamics for Improving Real-time Predictions in Future [73.03458424369657]
  In real-time forecasting in public health, data collection is a non-trivial and demanding task. 'Backfill' phenomenon and its effect on model performance has been barely studied in the prior literature. We formulate a novel problem and neural framework Back2Future that aims to refine a given model's predictions in real-time.
  arXiv  Detail & Related papers  (2021-06-08T14:48:20Z)
• Churn Reduction via Distillation [54.5952282395487]
  We show an equivalence between training with distillation using the base model as the teacher and training with an explicit constraint on the predictive churn. We then show that distillation performs strongly for low churn training against a number of recent baselines.
  arXiv  Detail & Related papers  (2021-06-04T18:03:31Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.