Towards Hierarchical Task Decomposition using Deep Reinforcement Learning for Pick and Place Subtasks
This work aims to reduce the data requirements for training robotic pick and place policies, which is a significant practical bottleneck for robotics researchers and developers.
This paper addresses the data inefficiency of Deep Reinforcement Learning (DRL) and Learning from Demonstrations (LfD) for robotic pick and place tasks by proposing a multi-subtask reinforcement learning methodology. They decompose complex tasks into low-level subtasks, train them with DRL, and combine them with a high-level choreographer, demonstrating improved sample-efficiency over LfD in simulation and robust grasping on a real robot.
Deep Reinforcement Learning (DRL) is emerging as a promising approach to generate adaptive behaviors for robotic platforms. However, a major drawback of using DRL is the data-hungry training regime that requires millions of trial and error attempts, which is impractical when running experiments on robotic systems. Learning from Demonstrations (LfD) has been introduced to solve this issue by cloning the behavior of expert demonstrations. However, LfD requires a large number of demonstrations that are difficult to be acquired since dedicated complex setups are required. To overcome these limitations, we propose a multi-subtask reinforcement learning methodology where complex pick and place tasks can be decomposed into low-level subtasks. These subtasks are parametrized as expert networks and learned via DRL methods. Trained subtasks are then combined by a high-level choreographer to accomplish the intended pick and place task considering different initial configurations. As a testbed, we use a pick and place robotic simulator to demonstrate our methodology and show that our method outperforms a benchmark methodology based on LfD in terms of sample-efficiency. We transfer the learned policy to the real robotic system and demonstrate robust grasping using various geometric-shaped objects.