GONet: A Semi-Supervised Deep Learning Approach For Traversability Estimation
This work addresses the practical challenge of safe robot navigation in indoor environments by reducing the need for costly negative data collection, though it is incremental as it builds on existing GAN-based methods.
The authors tackled the problem of traversability estimation for robots using fisheye images by proposing GONet, a semi-supervised deep learning method based on GANs that requires few negative examples, and demonstrated its robustness, generalization, and real-time capability on new datasets.
We present semi-supervised deep learning approaches for traversability estimation from fisheye images. Our method, GONet, and the proposed extensions leverage Generative Adversarial Networks (GANs) to effectively predict whether the area seen in the input image(s) is safe for a robot to traverse. These methods are trained with many positive images of traversable places, but just a small set of negative images depicting blocked and unsafe areas. This makes the proposed methods practical. Positive examples can be collected easily by simply operating a robot through traversable spaces, while obtaining negative examples is time consuming, costly, and potentially dangerous. Through extensive experiments and several demonstrations, we show that the proposed traversability estimation approaches are robust and can generalize to unseen scenarios. Further, we demonstrate that our methods are memory efficient and fast, allowing for real-time operation on a mobile robot with single or stereo fisheye cameras. As part of our contributions, we open-source two new datasets for traversability estimation. These datasets are composed of approximately 24h of videos from more than 25 indoor environments. Our methods outperform baseline approaches for traversability estimation on these new datasets.