Accelerating Grasp Learning via Pretraining with Coarse Affordance Maps of Objects

TL;DR

This paper introduces a pretraining approach using coarse affordance maps to significantly accelerate robotic grasp learning, reducing training time and improving success rates in simulation and real-world scenarios.

Contribution

It proposes a novel pretraining method with coarse affordance maps that greatly speeds up grasp learning and reduces labeling effort with minimal data.

Findings

Accelerates grasp learning by nearly three times in simulation.

Achieves 89.5% grasp success rate on a real robot with only 500 tries.

Demonstrates strong generalization to unseen objects.

Abstract

Self-supervised grasp learning, i.e., learning to grasp by trial and error, has made great progress. However, it is still time-consuming to train such a model and also a challenge to apply it in practice. This work presents an accelerating method of robotic grasp learning via pretraining with coarse affordance maps of objects to be grasped based on a quite small dataset. A model generated through pre-training is harnessed as an initialization policy to warmly start grasp learning so as to guide a robot to capture more effective rewards at the beginning of training. An object in its coarse affordance map is annotated with a single key point and thereby, the burden of labeling is greatly alleviated. Extensive experiments in simulation and on a real robot are conducted to evaluate the proposed method. The simulation results show that it can significantly accelerate grasp learning by nearly three times over a vanilla Deep Q-Network -based method. Its test on a real UR3 robot shows that it reaches a grasp success rate of 89.5% via only 500 times of grasp tries within about two hours, which is four times faster than its competitor. In addition, it enjoys an outstanding generalization ability to grasp prior-unseen novel objects. It outperforms some existing methods and has the potential to directly apply to a robot for real-world grasp learning tasks.