ML-based controller enhances quadrotor trajectory tracking

Researchers have developed a novel machine learning-based feedback-linearization control framework for quadrotors, designed to handle unmodeled dynamics and nonlinearities. This system utilizes a Gaussian Radial Basis Function neural network that updates its weights in real time to compensate for uncertainties like air drag and actuator dynamics. The control law is theoretically guaranteed to ensure closed-loop stability and asymptotic convergence for trajectory tracking. Experiments on the Crazyflie 2.1 quadrotor demonstrated significant improvements in tracking accuracy, reducing position-norm and yaw orientation RMSE by over 7.13% and 49.27% respectively, compared to a baseline controller. AI

IMPACT This research could lead to more robust and precise autonomous navigation for drones in complex and unpredictable environments.

RANK_REASON Academic paper detailing a new control method for quadrotors. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• neural network
• Bitcraze
• Crazyflie 2.1

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →