This study proposes a learning-based method with domain adaptability for input estimation of vehicle suspension systems. In a crowdsensing setting for bridge health monitoring, vehicles carry sensors to collect samples of the bridge’s dynamic response. The primary challenge is in preprocessing; signals are highly contaminated from road profile roughness and vehicle suspension dynamics. Additionally, signals are collected from a diverse set of vehicles vitiating model-based approaches. In our data-driven approach, two autoencoders for the cabin signal and the tire-level signal are constrained to force the separation of the tire-level input from the suspension system in the latent state representation. From the extracted features, we estimate the tire-level signal and deter-mine the vehicle class with high accuracy (98% classification accuracy). Compared to existing solutions for the vehicle suspension deconvolution problem, we show that the proposed methodology is robust to vehicle dynamic variations and suspension system nonlinearity.

Discovering the underlying behavior of complex systems is an important topic in many science and engineering disciplines. In this paper, we propose a novel neural network framework, finite difference neural networks (FD-Net), to learn partial differential equations from data. Specifically, our proposed finite difference inspired network is designed to learn the underlying governing partial differential equations from trajectory data, and to iteratively estimate the future dynamical behavior using only a few trainable parameters. We illustrate the performance (predictive power) of our framework on the heat equation, with and without noise and/or forcing, and compare our results to the Forward Euler method. Moreover, we show the advantages of using a Hessian-Free Trust Region method to train the network.

Cities are encountering extensive deficits in infrastructure service while they are experiencing rapid technological advancements and overhauls in transportation systems. Standard bridge evaluation methods rely on visual inspections, which are infrequent and subjective, ultimately affecting the structural assessments on which maintenance plans are based. The operational behavior of a bridge must be observed more regularly and over an extended period in order to sufficiently track its condition and avoid unexpected rehabilitation. Mobile sensor networks are conducive to monitoring bridges vibrations routinely, with benefits that have been demonstrated in recent structural health monitoring (SHM) research. Though smartphone accelerometers are imperfect sensors, they can contribute valuable information to SHM, especially when aggregated, e.g., via crowdsourcing. In an application on the Harvard Bridge (Boston, MA), it is shown that acceleration data collected using smartphones in moving vehicles contained consistent and significant indicators of the first three modal frequencies of the bridge. In particular, the results became more precise when informatics from several smartphone datasets were combined. This evidence is the first to support the hypothesis that smartphone data, collected within vehicles passing over a bridge, can be used to detect several modal frequencies of the bridge. The result defines an opportunity for local governments to make partnerships that encourage the collection of low-cost bridge vibration data, which can contribute to more effective management and informed decision-making.