Real World Robotic Exploration using Deep Neural Networks Trained in Photorealistic Reconstructed Environments
This work provides a robust navigation pipeline for real-world indoor robotics, though it is incremental as it builds on an existing deep learning approach.
The authors tackled robot localization in indoor scenes by modifying a deep neural network's loss function to combine positional and rotational errors, achieving reductions of up to 9.64% in median positional error and 2.99% in rotational error compared to the baseline, and developed a pipeline for real-time navigation with accuracies of 0.11m and 0.89 degrees.
In this work, an existing deep neural network approach for determining a robot's pose from visual information (RGB images) is modified, improving its localization performance without impacting its ease of training. Explicitly, the network's loss function is extended in a manner which intuitively combines the positional and rotational error in order to increase robustness to perceptual aliasing. An improvement in the localization accuracy for indoor scenes is observed: with decreases of up to 9.64% and 2.99% in the median positional and rotational error respectively, when compared to the unmodified network. Additionally, photogrammetry data is used to produce a pose-labelled dataset which allows the above model to be trained on a local environment, resulting in localization accuracies of 0.11m & 0.89 degrees. This trained model forms the basis of a navigation algorithm, which is tested in real-time on a TurtleBot (a wheeled robotic device). As such, this work introduces a full pipeline for creating a robust navigational algorithm for any given real world indoor scene; the only requirement being a collection of images from the scene, which can be captured in as little as 330 seconds of