KNODE-MPC: A Knowledge-based Data-driven Predictive Control Framework for Aerial Robots. (arXiv:2109.04821v3 [cs.RO] UPDATED)

In this work, we consider the problem of deriving and incorporating accurate
dynamic models for model predictive control (MPC) with an application to
quadrotor control. MPC relies on precise dynamic models to achieve the desired
closed-loop performance. However, the presence of uncertainties in complex
systems and the environments they operate in poses a challenge in obtaining
sufficiently accurate representations of the system dynamics. In this work, we
make use of a deep learning tool, knowledge-based neural ordinary differential
equations (KNODE), …

arxiv data data-driven framework predictive robots