Search for: All records

Purpose This research introduces an innovative solution that revolutionizes the study of linear and nonlinear dynamical systems—a smart automatic modal hammer. With its affordability and intelligent capabilities, this automatic modal hammer becomes an invaluable tool for research and industry, enabling repeatable strikes with precise force control. This system's significance becomes particularly evident when studying nonlinear systems, which heavily rely on the excitation level for their dynamics. By offering a cost-effective design this proposed system proves to be robust in accelerating research on nonlinear dynamics, providing researchers with an efficient and accessible means to delve deeper into these complex systems. Methods The proposed design integrates a commercial modal hammer, commonly used in modal testing, and a stepper motor. This stepper motor is enhanced with an encoder and servo driver, all expertly controlled by a Raspberry Pi. Results What sets this system apart is its clever utilization of regression models to acquire knowledge of the intrinsic relationship between the applied force and hammer velocity precisely during the impact. This acquired knowledge is the foundation for controlling the motor's behavior, ensuring consistent and accurate excitation of the structure with the desired force. Conclusion The capabilities of the proposed automatic modal hammer are demonstrated using a linear two-story tower and a model airplane wing with a nonlinear vibration absorber. 

In the context of medical artificial intelligence, this study explores the vulnerabilities of the Pathology Language-Image Pretraining (PLIP) model, a Vision Language Foundation model, under targeted attacks. Leveraging the Kather Colon dataset with 7,180 H&E images across nine tissue types, our investigation employs Projected Gradient Descent (PGD) adversarial perturbation attacks to induce misclassifications intentionally. The outcomes reveal a 100% success rate in manipulating PLIP’s predictions, underscoring its susceptibility to adversarial perturbations. The qualitative analysis of adversarial examples delves into the interpretability challenges, shedding light on nuanced changes in predictions induced by adversarial manipulations. These findings contribute crucial insights into the interpretability, domain adaptation, and trustworthiness of Vision Language Models in medical imaging. The study emphasizes the pressing need for robust defenses to ensure the reliability of AI models. The source codes for this experiment can be found at https://github.com/jaiprakash1824/VLM Adv Attack.