Smart Control for Precision Mechanical Weed Management on a Mobile Robot Platform

Mechanical weed management is a major bottleneck for sustainable yet efficient food production. Intelligent and versatile intra-row weeding solutions remain unavailable due to the high complexity of agricultural environments. This project aims to create a learning methodology that combines model-based learning of dynamics with model-free residual reinforcement learning in a simulated environment to learn adaptive control policies for weed management.