Learning to Fly in Seconds

Keyphrases

• Reinforcement Learning 100%
• Training Time 100%
• Laptop 66%
• Learning-based 66%
• Fast Training 66%
• Simulation to Real 66%
• Highly Reliable 33%
• Nonlinearity 33%
• Performance Enhancement 33%
• Control Solution 33%
• Microcontroller 33%
• Existing State 33%
• Trajectory Tracking 33%
• Sample Complexity 33%
• Barriers to Entry 33%
• Training Paradigm 33%
• Competitive Performance 33%
• Quadrotor 33%
• Control Complex 33%
• Multirotor 33%
• Quadrotor Control 33%
• Consumer-grade 33%
• Deep Reinforcement Learning (deep RL) 33%
• Multi-rotor Aerial Vehicles 33%
• Curriculum Learning 33%
• Real-time Guarantees 33%
• Actor-critic 33%
• Dynamic Simulator 33%
• Sim-to-real Transfer 33%
• Crazyflie 33%
• Reality Gap 33%
• Revolutions per Minute 33%
• Multirotor Control 33%
• Multirotor Dynamics 33%
• Nano-quadrotor 33%

Computer Science

• Reinforcement Learning 100%
• Open Source 33%
• Microcontroller 33%
• Deep Reinforcement Learning 33%
• Control Abstraction 33%
• Graphics Processing Unit 33%