Neural Lander: Stable Drone Landing Control Using Learned Dynamics

By Guanya Shi, Xichen Shi, Michael O'Connell, Rose Yu, Kamyar Azizzadenesheli, Anima Anandkumar, Yisong Yue, Soon-Jo Chung. IEEE International Conference on Robotics and Automation, 2018

350 citations recorded by the source corpus.

Precise near-ground trajectory control is difficult for multi-rotor drones, due to the complex aerodynamic effects caused by interactions between multi-rotor airflow and the environment. Conventional control methods often fail to properly account for these complex effects and fall short in accomplishing smooth landing. In this paper, we present a novel deep-learning-based robust nonlinear controller (Neural-Lander) that improves control performance of a quadrotor during landing. Our approach combines a nominal dynamics model with a Deep Neural Network (DNN) that learns high-order interactions. We apply spectral normalization (SN) to constrain the Lipschitz constant of the DNN. Leveraging this Lipschitz property, we design a nonlinear feedback linearization controller using the learned model and prove system stability with disturbance rejection. To the best of our knowledge, this is the first DNN-based nonlinear feedback controller with stability guarantees that can utilize arbitrarily large neural nets. Experimental results demonstrate that the proposed controller significantly outperforms a Baseline Nonlinear Tracking Controller in both landing and cross-table trajectory tracking cases. We also empirically show that the DNN generalizes well to unseen data outside the training domain.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/204bcdbb217a96333279ad307aeb1d0729365832

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

Citations in the corpus

Meta-Adaptive Nonlinear Control: Theory and Algorithms (2021, methodology): “We use a DNN with spectral normalization [38, 25, 39, 40] to represent φ, and use the Adam optimizer [41] for A1.”
Neural-Fly enables rapid learning for agile flight in strong winds (2022, methodology): “prediction error term + P  T s ⏟ tracking error term (8)”
Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning (2021, methodology): “Learning Nonparametric Unknown Dynamics with Machine Learning Models.”
Safe Control With Learned Certificates: A Survey of Neural Lyapunov, Barrier, and Contraction Methods for Robotics and Control (2022, background): “Some disturbances, such as unmodeled aerodynamic effects [80], can be represented as unknown forces added to the known system dynamics, while others, such as uncertain mass or inertia, must be modeled as potentially multiplicative disturbances [10].”