A Reinforcement Learning Approach for Robotic Unloading from Visual Observations
This work addresses the challenge of autonomous robotic unloading in realistic scenarios where labeled data is scarce, offering a domain-specific solution.
The paper tackles the problem of robotic unloading from visual observations by developing a sample-efficient hierarchical controller that combines deep reinforcement learning with classical motion control, achieving improved learning performance without labeled data.
In this work, we focus on a robotic unloading problem from visual observations, where robots are required to autonomously unload stacks of parcels using RGB-D images as their primary input source. While supervised and imitation learning have accomplished good results in these types of tasks, they heavily rely on labeled data, which are challenging to obtain in realistic scenarios. Our study aims to develop a sample efficient controller framework that can learn unloading tasks without the need for labeled data during the learning process. To tackle this challenge, we propose a hierarchical controller structure that combines a high-level decision-making module with classical motion control. The high-level module is trained using Deep Reinforcement Learning (DRL), wherein we incorporate a safety bias mechanism and design a reward function tailored to this task. Our experiments demonstrate that both these elements play a crucial role in achieving improved learning performance. Furthermore, to ensure reproducibility and establish a benchmark for future research, we provide free access to our code and simulation.