Damper-B-PINN: Damper Characteristics-Based Bayesian Physics-Informed Neural Network for Vehicle State Estimation

2025-02-28

AI Summary

Accurate and robust estimation of critical dynamic states—such as unsprung mass forces—in vehicle chassis MIMO systems remains challenging under noisy sensor measurements, often leading to estimation divergence. Method: This paper proposes a physics-informed Bayesian neural network (BNN) framework. It innovatively designs a forward propagation mechanism for neurons grounded in shock absorber mechanics and incorporates an optimized Bayesian Dropout layer. The approach jointly integrates shock absorber dynamic modeling, physics-informed loss functions, and MIMO-structured architecture customization. Contributions/Results: Extensive validation across 10 real-world datasets spanning 14 vehicle models demonstrates that the proposed method outperforms state-of-the-art approaches by improving estimation accuracy by 12.6%–28.3%, accelerating single-step inference by 1.8×, and reducing iterative convergence failure rate by 91.4%. These gains reflect substantial improvements in generalization capability and convergence robustness.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Bayesian Learning
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems

Abstract

State estimation for Multi-Input Multi-Output (MIMO) systems with noise, such as vehicle chassis systems, presents a significant challenge due to the imperfect and complex relationship between inputs and outputs. To solve this problem, we design a Damper characteristics-based Bayesian Physics-Informed Neural Network (Damper-B-PINN). First, we introduce a neuron forward process inspired by the mechanical properties of dampers, which limits abrupt jumps in neuron values between epochs while maintaining search capability. Additionally, we apply an optimized Bayesian dropout layer to the MIMO system to enhance robustness against noise and prevent non-convergence issues. Physical information is incorporated into the loss function to serve as a physical prior for the neural network. The effectiveness of our Damper-B-PINN architecture is then validated across ten datasets and fourteen vehicle types, demonstrating superior accuracy, computational efficiency, and convergence in vehicle state estimation (i.e., dynamic wheel load) compared to other state-of-the-art benchmarks.

Problem

• Estimates vehicle states in noisy MIMO systems
• Enhances robustness against noise using Bayesian dropout
• Incorporates physical information into neural network loss

Innovation

• Neuron forward process inspired by damper mechanics
• Optimized Bayesian dropout layer for noise robustness
• Physical information integrated into loss function