Rough surface lubrication simulation is crucial for designing and optimizing tribological performance. Despite the growing application of Physical Information Neural Networks (PINNs) in hydrodynamic lubrication analysis, their use has been primarily limited to smooth surfaces. This is due to traditional PINN methods suffer from spectral bias, favoring to learn low-frequency features and thus failing to analyze rough surfaces with high-frequency signals. To date, no PINN methods have been reported for rough surface lubrication. To overcome these limitations, this work introduces a novel multi-scale lubrication neural network architecture that utilizes a trainable Fourier feature network. By incorporating learnable feature embedding frequencies, this architecture automatically adapts to various frequency components, thereby enhancing the analysis of rough surface characteristics. This method has been tested across multiple surface morphologies, and the results have been compared with those obtained using the finite element method (FEM). The comparative analysis demonstrates that this approach achieves a high consistency with FEM results. Furthermore, this novel architecture surpasses traditional Fourier feature networks with fixed feature embedding frequencies in both accuracy and computational efficiency. Consequently, the multi-scale lubrication neural network model offers a more efficient tool for rough surface lubrication analysis.

Have you ever lost your grip on something that you've dropped into the swimming pool, or worse, toilet? Scientists may have developed a solution to holding onto underwater objects, but it is not primarily intended to help you rescue your iPhone from a watery fate. Researchers at Virginia Tech have developed a glove that will allow divers to get a firm grasp while, for example, rescuing someone or salvaging a shipwreck. The'octa-glove' is inspired by octopus tentacles, and is covered in robotic suckers equipped with sensors that can tell how far away an object is. When the sensors detect a nearby surface, it sends a signal to the controller which will activate the sucker's adhesion.

Over a decade ago, Stanford roboticists started experimenting with ways of using arrays of very small spines to help climbing robots grip rough surfaces. These microspine grippers have been used on all kinds of research robots since then, and recently, NASA has decided that microspines are the best way for spacecraft to grab onto asteroids. Yesterday at the IEEE/RSJ International Conference on Intelligent Robots and Systems in South Korea, Shiquan Wang from Stanford presented a new microspine-based palm design for rock-climbing robots. These palms use microspines that can support four times the weight of previous designs, which will be enough to turn JPL's RoboSimian DRC robot into a champion rock climber. And we're not talking just scrambling up slopes: It'll be able to scale vertical rock faces, and even clamber around overhangs.