Using meta-reasoning for incremental repairs in multi-object robot manipulation tasks

Robots tasked with object assembly by manipulation of parts require not only a high-level plan for order placement but also detailed low-level information on how to place and pick the part based its state. This is complex multi-level problem prone failures at various levels. paper employs meta reasoning architecture along robotics principles proposes dual encoding state expectations during progression task ground nominal scenarios. We present our results table-top scenario using perceptual in concept occupancy grids key point representations. Our constrained setting suggest or alone system performs worse than if uses them both. then outline complete which tackles this repairing more generic plans objects moving spaces 6 degrees freedom.