Design of Deep Neural Networks as Add-on Blocks for Improving Impromptu Trajectory Tracking

This paper provides guidelines for designing deep neural networks (DNNs) as add-on blocks to baseline feedback control loops to enhance tracking performance on arbitrary, desired trajectories. The DNNs are trained to adapt the reference signals to the feedback control loop. The goal is to achieve a unity map between the desired and actual outputs. In previous work, the efficacy of this approach was demonstrated on quadrotors. On 30 unseen trajectories, the proposed DNN approach achieved an average error reduction of 43%, compared to the baseline feedback controller. Motivated by these results, this work aims to provide platform-independent design guidelines for the proposed DNN-enhanced control architecture. In particular, we provide specific guidelines for the DNN feature selection, derive conditions for when the proposed approach is effective, and show in which cases training efficiency can be further increased.