Caris Moses

Papers

Total Citations

H-Index

About

Caris Moses is a rising researcher in robotics and artificial intelligence, with a focus on long-horizon sequential manipulation and hierarchical planning. Their most-cited work, "Active Learning of Abstract Plan Feasibility" (2021, 13 citations), tackles a critical bottleneck in robotics: how to efficiently predict whether a high-level abstract plan can be successfully executed by lower-level motion planners. By introducing an active learning framework that selectively queries the most informative abstract actions, Moses enables robots to avoid costly, failed motion planning attempts—making hierarchical task-and-motion planning more practical and scalable. This contribution bridges the gap between symbolic reasoning and geometric feasibility, a key challenge in autonomous manipulation. Though early in their career, Moses’s work demonstrates a clear impact on the field, offering a data-driven path to more robust robot autonomy. Their research is particularly relevant for students and engineers working on robot manipulation, planning under uncertainty, and learning for control—areas where efficient decision-making is paramount.