Zhang, YunboClegg, AlexanderHa, SehoonTurk, GregYe, YutingMyszkowski, KarolNiessner, Matthias2023-05-032023-05-0320231467-8659https://doi.org/10.1111/cgf.14741https://diglib.eg.org:443/handle/10.1111/cgf14741In-hand object manipulation is challenging to simulate due to complex contact dynamics, non-repetitive finger gaits, and the need to indirectly control unactuated objects. Further adapting a successful manipulation skill to new objects with different shapes and physical properties is a similarly challenging problem. In this work, we show that natural and robust in-hand manipulation of simple objects in a dynamic simulation can be learned from a high quality motion capture example via deep reinforcement learning with careful designs of the imitation learning problem. We apply our approach on both single-handed and two-handed dexterous manipulations of diverse object shapes and motions. We then demonstrate further adaptation of the example motion to a more complex shape through curriculum learning on intermediate shapes morphed between the source and target object. While a naive curriculum of progressive morphs often falls short, we propose a simple greedy curriculum search algorithm that can successfully apply to a range of objects such as a teapot, bunny, bottle, train, and elephant.CCS Concepts: Computing methodologies -> Physical simulation; Motion capture; Reinforcement learning; Learning from demonstrationsComputing methodologiesPhysical simulationMotion captureReinforcement learningLearning from demonstrations10.1111/cgf.1474125-3612 pages