A Robust Adaptive Controller for a Seed Refilling System on a Moving Platform

This paper considers an opportunity for robotic automation in the seeding of field crops. It reports on progress in designing a partially automated seed refilling system that transfers seeds to a centralized tank on a tractor-pulled planter during travel, while the tractor continues to follow a trajectory determined by a desired seeding strategy The paper describes challenges associated with accommodating variability among commercial planters, as well as with formulating a control strategy for a robotic manipulator that compensates for time-dependent uncertainty and unknown disturbance loads due to travel of a mounting platform across uneven terrain. To address such challenges, the paper reviews theoretical predictions for a recently proposed adaptive control framework regarding the existence of transient performance bounds in the presence of fast uncertainty, as well as robustness to unmodeled dynamics in the form of time delays in the control signals. In particular, the theoretical treatment guarantees the existence of a positive time-delay margin, below which stable operation can be expected. Several experimental designs involving one or multiple robotic manipulators are considered in order to validate the theoretical results and to explore possible seed transfer strategies.