Towards bio-inspired unsupervised representation learning for indoor aerial navigation

TL;DR

This paper introduces a biologically inspired unsupervised deep learning approach for indoor drone navigation that produces low-dimensional representations, enabling robust SLAM in GPS-denied environments on energy-efficient hardware.

Contribution

It presents a novel unsupervised representation learning method inspired by biological systems for drone SLAM in indoor environments, optimized for embedded hardware.

Findings

Feasibility demonstrated on indoor warehouse dataset

Low-dimensional latent descriptors improve robustness

Suitable for power-constrained embedded systems

Abstract

Aerial navigation in GPS-denied, indoor environments, is still an open challenge. Drones can perceive the environment from a richer set of viewpoints, while having more stringent compute and energy constraints than other autonomous platforms. To tackle that problem, this research displays a biologically inspired deep-learning algorithm for simultaneous localization and mapping (SLAM) and its application in a drone navigation system. We propose an unsupervised representation learning method that yields low-dimensional latent state descriptors, that mitigates the sensitivity to perceptual aliasing, and works on power-efficient, embedded hardware. The designed algorithm is evaluated on a dataset collected in an indoor warehouse environment, and initial results show the feasibility for robust indoor aerial navigation.