Synthetic Data Generation for Minimum-Exposure Navigation in a Time-Varying Environment using Generative AI Models
This work addresses data scarcity in autonomous navigation by generating synthetic threat fields, but it is incremental as it builds on existing generative and recurrent models with a novel split latent space approach.
The paper tackles the problem of generating synthetic environmental threat field data for autonomous vehicle navigation when real-world data is scarce, proposing a split variational recurrent neural network (S-VRNN) that merges variational autoencoders and recurrent neural networks to produce statistically similar samples, with numerical experiments showing effectiveness even with very small training datasets.
We study the problem of synthetic generation of samples of environmental features for autonomous vehicle navigation. These features are described by a spatiotemporally varying scalar field that we refer to as a threat field. The threat field is known to have some underlying dynamics subject to process noise. Some "real-world" data of observations of various threat fields are also available. The assumption is that the volume of ``real-world'' data is relatively small. The objective is to synthesize samples that are statistically similar to the data. The proposed solution is a generative artificial intelligence model that we refer to as a split variational recurrent neural network (S-VRNN). The S-VRNN merges the capabilities of a variational autoencoder, which is a widely used generative model, and a recurrent neural network, which is used to learn temporal dependencies in data. The main innovation in this work is that we split the latent space of the S-VRNN into two subspaces. The latent variables in one subspace are learned using the ``real-world'' data, whereas those in the other subspace are learned using the data as well as the known underlying system dynamics. Through numerical experiments we demonstrate that the proposed S-VRNN can synthesize data that are statistically similar to the training data even in the case of very small volume of ``real-world'' training data.