Vision-Based Autonomous Drone Control using Supervised Learning in Simulation

TL;DR

This paper presents a vision-based control system for autonomous micro aerial vehicles using supervised learning in simulation, enabling navigation and landing in GPS-denied indoor environments with limited resources.

Contribution

It introduces a CNN-based approach trained on simulated data for MAV control, reducing training time compared to reinforcement learning and manual data collection.

Findings

Successful navigation in simulated environments

Effective landing on designated platforms

Shorter training times than reinforcement learning methods

Abstract

Limited power and computational resources, absence of high-end sensor equipment and GPS-denied environments are challenges faced by autonomous micro areal vehicles (MAVs). We address these challenges in the context of autonomous navigation and landing of MAVs in indoor environments and propose a vision-based control approach using Supervised Learning. To achieve this, we collected data samples in a simulation environment which were labelled according to the optimal control command determined by a path planning algorithm. Based on these data samples, we trained a Convolutional Neural Network (CNN) that maps low resolution image and sensor input to high-level control commands. We have observed promising results in both obstructed and non-obstructed simulation environments, showing that our model is capable of successfully navigating a MAV towards a landing platform. Our approach requires shorter training times than similar Reinforcement Learning approaches and can potentially overcome the limitations of manual data collection faced by comparable Supervised Learning approaches.