Task Planning on Stochastic Aisle Graphs for Precision Agriculture

This work addresses task planning under uncertainty for precision agriculture applications whereby costs are uncertain and the gain of completing a is proportional to resource consumption (such as water in irrigation). The goal complete all tasks while prioritizing those that more urgent, subject diverse budget thresholds stochastic tasks. To describe agriculture-related environments incorporate tasks, new Stochastic-Vertex-Cost Aisle Graph (SAG) introduced. Then, allocation algorithm, termed Next-Best-Action Planning (NBA-P), proposed. NBA-P utilizes underlying structure enabled by SAG, tackles problem simultaneously determining optimal perform an time exit (i.e. return base station), at run-time. proposed approach tested with both simulated data real-world experimental datasets collected commercial vineyard, single- multi-robot scenarios. In cases, outperforms other evaluated methods terms per visited vertex, wasted resources resulting from aborted when threshold exceeded), total vertices.