English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Modeling the precise dynamics of off-road vehicles is a complex yet essentialtask due to the challenging terrain they encounter and the need for optimalperformance and safety. Recently, there has been a focus on integrating nominalphysics-based models alongside data-driven neural networks using PhysicsInformed Neural Networks. These approaches often assume the availability of awell-distributed dataset; however, this assumption may not hold due to regionsin the physical distribution that are hard to collect, such as high-speedmotions and rare terrains. Therefore, we introduce a physics-informed dataaugmentation methodology called PIAug. We show an example use case of the sameby modeling high-speed and aggressive motion predictions, given a dataset withonly low-speed data. During the training phase, we leverage the nominal modelfor generating target domain (medium and high velocity) data using theavailable source data (low velocity). Subsequently, we employ aphysics-inspired loss function with this augmented dataset to incorporate priorknowledge of physics into the neural network. Our methodology results in up to67% less mean error in trajectory prediction in comparison to a standalonenominal model, especially during aggressive maneuvers at speeds outside thetraining domain. In real-life navigation experiments, our model succeeds in 4xtighter waypoint tracking constraints than the Kinematic Bicycle Model (KBM) atout-of-domain velocities.

PIAug -- Physics Informed Augmentation for Learning Vehicle Dynamics for Off-Road Navigation