Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This study explores the application of conditional invertible neural networks (cINNs) as probabilistic models for inverse dynamics in multirotor UAV control, specifically focusing on a planar X8 coaxial multicopter. By leveraging an incremental nonlinear dynamic inversion (INDI) teacher and employing rational-quadratic spline coupling, the model achieves high open-loop performance with an R² of 0.944 and a mean CRPS of 0.0915. In closed-loop scenarios, the cINN's position RMSE closely matches that of the INDI method, highlighting the model's effectiveness while identifying command bandwidth and data coverage as critical factors influencing performance failures.
Achieving near-parity with traditional methods, cINNs reveal critical insights into command bandwidth limitations in UAV control.
We investigate conditional invertible neural networks (cINNs) as probabilistic inverse-dynamics models for multirotor control. For a planar X8 coaxial multicopter, we learn $p(u \mid s_t, c_t)$ from an incremental nonlinear dynamic inversion (INDI) teacher using rational-quadratic spline coupling and invertible linear mixing. Open-loop reproduction reaches $R^2 = 0.944$, mean CRPS 0.0915, and log-probability-error correlation $\rho = -0.60$. Over 15 closed-loop scenarios, position RMSE matches INDI (9.7 vs. 9.5 m), with 47 percent tracking acceptably; failures separate into attitude divergence under aggressive steps and phase lag under high-frequency references, isolating command bandwidth and data coverage as dominant failure mechanisms.