PITA: Physics-Informed Trajectory Autoencoder

arXiv:2403.11728v1 Announce Type: new
Abstract: Validating robotic systems in safety-critical appli-cations requires testing in many scenarios including rare edgecases that are unlikely to occur, requiring to complement real-world testing with testing in simulation. Generative models canbe used to augment real-world datasets with generated data toproduce edge case scenarios by sampling in a learned latentspace. Autoencoders can learn said latent representation for aspecific domain by learning to reconstruct the input data froma lower-dimensional intermediate representation. However, theresulting trajectories are not necessarily …

abstract arxiv autoencoder autoencoders case cs.lg cs.ro data datasets edge generated generative generative models physics physics-informed robotic safety safety-critical sampling simulation systems testing trajectory type world