The scientific computation of large deformations in elastic-plastic solids is crucial in various manufacturing applications. Traditional numerical methods exhibit several inherent limitations, prompting Deep Learning (DL) as a promising alternative. The effectiveness of current DL techniques typically depends on the availability of high-quantity and high-accuracy datasets, which are yet difficult to obtain in large deformation problems. During the dataset construction process, a dilemma stands between data quantity and data accuracy, leading to suboptimal performance in the DL models. To address this challenge, we focus on a representative application of large deformations, the stretch bending problem, and propose FilDeep, a Fidelity-based Deep Learning framework for large Deformation of elastic-plastic solids. Our FilDeep aims to resolve the quantity-accuracy dilemma by simultaneously training with both low-fidelity and high-fidelity data, where the former provides greater quantity but lower accuracy, while the latter offers higher accuracy but in less quantity. In FilDeep, we provide meticulous designs for the practical large deformation problem. Particularly, we propose attention-enabled cross-fidelity modules to effectively capture long-range physical interactions across MF data. To the best of our knowledge, our FilDeep presents the first DL framework for large deformation problems using MF data. Extensive experiments demonstrate that our FilDeep consistently achieves state-of-the-art performance and can be efficiently deployed in manufacturing.

Multi-solid systems are foundational to a wide range of real-world applications, yet modeling their complex interactions remains challenging. Existing deep learning methods predominantly rely on implicit modeling, where the factors influencing solid deformation are not explicitly represented but are instead indirectly learned. However, as the number of solids increases, these methods struggle to accurately capture intricate physical interactions. In this paper, we introduce a novel explicit modeling paradigm that incorporates factors influencing solid deformation through structured modules. Specifically, we present Unisoma, a unified and flexible Transformer-based model capable of handling variable numbers of solids. Unisoma directly captures physical interactions using contact modules and adaptive interaction allocation mechanism, and learns the deformation through a triplet relationship. Compared to implicit modeling techniques, explicit modeling is more well-suited for multi-solid systems with diverse coupling patterns, as it enables detailed treatment of each solid while preventing information blending and confusion. Experimentally, Unisoma achieves consistent state-of-the-art performance across seven well-established datasets and two complex multi-solid tasks. Code is avaiable at [https://github.com/therontau0054/Unisoma](https://github.com/therontau0054/Unisoma).

Scientific computing for large deformation of elastic-plastic solids is critical for numerous real-world applications. Classical numerical solvers rely primarily on local discrete linear approximation and are constrained by an inherent trade-off between accuracy and efficiency. Recently, deep learning models have achieved impressive progress in solving the continuum mechanism. While previous models have explored various architectures and constructed coefficient-solution mappings, they are designed for general instances without considering specific problem properties and hard to accurately handle with complex elastic-plastic solids involving contact, loading and unloading. In this work, we take stretch bending, a popular metal fabrication technique, as our case study and introduce LaDEEP, a deep learning-based surrogate model for La rge De formation of Elastic-Plastic Solids. We encode the partitioned regions of the involved slender solids into a token sequence to maintain their essential order property. To characterize the physical process of the solid deformation, a two-stage Transformer-based module is designed to predict the deformation with the sequence of tokens as input. Empirically, LaDEEP achieves five magnitudes faster speed than finite element methods with a comparable accuracy, and gains 20.47% relative improvement on average compared to other deep learning baselines. We have also deployed our model into a real-world industrial production system, and it has shown remarkable performance in both accuracy and efficiency. Code is available at https://github.com/therontau0054/LaDEEP.

Accurate gradient and heavy vehicle mass estimation are crucial to improve vehicle control effects. However, significant obstacles remain in the way of accurately estimating gradient and mass in real-time, mainly reflected in the connection between vehicle mass and gradient. to instantly calculate the vehicle's mass and the gradient, this paper proposes a double variable forgetting factor recursive least squares(VFFRLS) algorithm based on vehicle longitudinal dynamics. Firstly, considering the deep coupling of mass and gradient, a multivariate VFFRLS algorithm is designed to obtain a joint mass-slope estimation, to determine the precise, steady vehicle mass and the uneven gradient. Secondly, because the slope of the road changes more frequently than the vehicle's mass, the upper layer estimation mass is used as input to design the lower layer VFFRLS algorithm to realize the fine estimation of road slope. The influence of cumulative error and coupling relationship on slope estimation accuracy is effectively reduced. Finally, To get an exact estimate of the slope of the road, the rough and fine estimates of the slope are distributed according to a specific proportion. The Trucksim and Matlab/Simulink simulations confirm the efficacy of this algorithm. The findings indicate that the real-time estimation algorithm designed can achieve a mass estimation error is less than 4%, and a slope estimation error is less than 3%.

The intelligent electric vehicle with wheel corner module not only exhibits exceptional mobility and stability, but also encounters the challenge of controlling excessive degrees of freedom. In order to enhance yaw stability, the yaw control method grounded in the concept of dynamic stability region is introduced. And the corresponding controllers are built based on the hierarchical control theory. The upper layer control uses the multi-constrained sideslip angle-yaw rate (β–w) phase curve to construct vehicle's dynamic stability region, the vehicle status is categorized into zones of absolute stability, transition and instability range. The stability margin which determines intervention time of the yaw control is designed according to the stability region. The middle layer control generates generalized force to sustain vehicle stability utilizing sliding mode algorithm. The bottom layer control is configured to allocate driving force of each wheel based on the wheel load. The proposed approach's effectiveness has been confirmed through a collaborative simulation involving MATLAB/Simulink and Carsim. The findings indicate that employing the yaw stability control method rooted in the dynamic stability region achieves a 55% reduction in sideslip angle error, thereby significantly enhancing yaw stability.