Neuroflight: Next Generation Flight Control Firmware

Abstract

Little innovation has been made to low-level attitude flight control used by uncrewed aerial vehicles (UAVs), which still predominantly uses the classical PID controller. In this work we introduce Neuroflight, the first open source neuro-flight controller firmware. We present our toolchain for training a neural network in simulation and compiling it to run on embedded hardware. Challenges faced jumping from simulation to reality are discussed along with our solutions. Our evaluation shows the neural network can execute at over 2.67kHz on an Arm Cortex-M7 processor and flight tests demonstrate a quadcopter running Neuroflight can achieve stable flight and execute aerobatic maneuvers.

References in corpus (2)

• Continuous control with deep reinforcement learning
• Trust Region Policy Optimization

Cited by in corpus (6)

• Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments
• Learning to Fly via Deep Model-Based Reinforcement Learning
• Flight Controller Synthesis Via Deep Reinforcement Learning
• Roofline Model for UAVs:A Bottleneck Analysis Tool for Designing Compute Systems for Autonomous Drones
• AutoPilot: Automating SoC Design Space Exploration for SWaP Constrained Autonomous UAVs
• How to Train your Quadrotor: A Framework for Consistently Smooth and Responsive Flight Control via Reinforcement Learning