Grasping in the Wild:Learning 6DoF Closed-Loop Grasping from Low-Cost Demonstrations
This addresses the challenge of flexible robotic manipulation for applications like logistics or home assistance, representing a novel method for a known bottleneck rather than an incremental improvement.
The paper tackles the problem of enabling robots to perform 6DoF closed-loop grasping in diverse real-world environments by proposing a low-cost hardware interface for collecting human demonstrations and training an end-to-end reinforcement learning model, achieving reliable grasping of novel objects across various static and dynamic scenes.
Intelligent manipulation benefits from the capacity to flexibly control an end-effector with high degrees of freedom (DoF) and dynamically react to the environment. However, due to the challenges of collecting effective training data and learning efficiently, most grasping algorithms today are limited to top-down movements and open-loop execution. In this work, we propose a new low-cost hardware interface for collecting grasping demonstrations by people in diverse environments. Leveraging this data, we show that it is possible to train a robust end-to-end 6DoF closed-loop grasping model with reinforcement learning that transfers to real robots. A key aspect of our grasping model is that it uses "action-view" based rendering to simulate future states with respect to different possible actions. By evaluating these states using a learned value function (Q-function), our method is able to better select corresponding actions that maximize total rewards (i.e., grasping success). Our final grasping system is able to achieve reliable 6DoF closed-loop grasping of novel objects across various scene configurations, as well as dynamic scenes with moving objects.