ReinforceGen: Hybrid Skill Policies with Automated Data Generation and Reinforcement Learning
This addresses the challenge of long-horizon manipulation for robotics, representing an incremental improvement through hybrid methods.
The paper tackles long-horizon manipulation in robotics by proposing ReinforceGen, a system that combines task decomposition, data generation, imitation learning, and reinforcement learning, achieving an 80% success rate on Robosuite tasks with visuomotor controls and an 89% average performance increase from fine-tuning.
Long-horizon manipulation has been a long-standing challenge in the robotics community. We propose ReinforceGen, a system that combines task decomposition, data generation, imitation learning, and motion planning to form an initial solution, and improves each component through reinforcement-learning-based fine-tuning. ReinforceGen first segments the task into multiple localized skills, which are connected through motion planning. The skills and motion planning targets are trained with imitation learning on a dataset generated from 10 human demonstrations, and then fine-tuned through online adaptation and reinforcement learning. When benchmarked on the Robosuite dataset, ReinforceGen reaches 80% success rate on all tasks with visuomotor controls in the highest reset range setting. Additional ablation studies show that our fine-tuning approaches contributes to an 89% average performance increase. More results and videos available in https://reinforcegen.github.io/