Easy-IIL: Reducing Human Operational Burden in Interactive Imitation Learning via an Assistant Expert

Easy-IIL is a framework that leverages off-the-shelf model-based imitation methods as an assistant expert to replace active human operation during data collection. The human expert provides a single demonstration to initialize the assistant and intervenes only in critical states. The method maintains IIL performance by preserving both offline and online data quality, as demonstrated through extensive simulation and real-world experiments.