The rapid development of autonomous driving has underscored the pivotal role of 3D perception. Monocular 3D object detection, as a cost-effective alternative to expensive lidar systems, is garnering increasing attention. However, existing pseudo-lidar methods encounter challenges such as coarse quality and insufficient semantic information when generating 3D point clouds from monocular images. To address these issues, this paper introduces MSFNet3D, which aims to overcome the quality limitations of pseudo-lidar point cloud. Our contributions are threefold: (1) We introduce a dual-branch network to optimize depth maps and propose a multi-scale channel spatial attention module (MS_CBAM). This module captures multi-scale geometric features through a hierarchical feature pyramid and an adaptive weight allocation mechanism, thereby addressing the scale sensitivity inherent in traditional attention mechanisms. (2) We propose a consistency-weighted fusion strategy that employs local gradient consistency analysis and differentiable weighted optimization to achieve a pixel-level fusion of image and depth features. This approach reduces feature conflicts within the dual-branch network and enhances the model’s robustness in complex scenes. (3) We introduce a semantic-guided pseudo-point cloud enhancement method that leverages an instance segmentation network to extract object-specific semantic regions and generate high-confidence point cloud, consequently improving the accuracy of object detection. Experiments on the KITTI dataset show that the proposed method performs excellently under various detection challenges, achieving an average precision of 18.87% in the 3D detection of car objects, which is a 1.67% improvement over the original model. The method also shows good performance in detecting pedestrians and cyclists. The proposed framework can provide economical and reliable 3D perception for mass-produced electric vehicles.

Accurate online tracking of state of charge (SOC) inconsistency in lithium-ion battery packs is essential for safety. It is equally critical for effective battery management in real-world operation. To achieve robust performance in dynamic battery environments characterized by temperature fluctuations and cell aging, a method combining enhanced Rime optimized-unscented particle filter (ERIME-UPF) with cubature smooth variable structure filter-varying boundary layer (CSVSF-VBL) is proposed. The cell mean-difference model is used to simulate the behavior characteristics of the battery module, including the hysteresis effect dynamic migration model, and the Rint model. First, module SOC is estimated using an ERIME-UPF, which adaptively adjusts the noise covariances of UPF via the enhanced RIME optimizer. Simultaneously, CSVSF-VBL employs the Rint model to estimate cell SOC inconsistencies, incorporating capacity and internal resistance coefficients into the second-order performance chattering to better capture cell inconsistency. Experiments focus on LiFePO₄ batteries under various inconsistencies, temperature, and aging states. The results show that ERIME-UPF achieves an average mean absolute error (MAE) of 0.33% for module SOC estimation, while CSVSF-VBL achieves a peak MAE of 3.28% for cell SOC estimation. Demonstrating superior accuracy and robustness in tracking SOC inconsistency under dynamic and degraded operating conditions.