Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping
Autor: | Alex Irpan, Sergey Levine, Vincent Vanhoucke, Konstantinos Bousmalis, Paul Wohlhart, Peter Pastor, Laura Downs, Julian Ibarz, Yunfei Bai, Matthew Kelcey, Mrinal Kalakrishnan, Kurt Konolige |
---|---|
Rok vydání: | 2017 |
Předmět: |
FOS: Computer and information sciences
0209 industrial biotechnology Computer Science - Artificial Intelligence Computer science business.industry Computer Vision and Pattern Recognition (cs.CV) GRASP Feature extraction Computer Science - Computer Vision and Pattern Recognition 02 engineering and technology Term (time) Machine Learning (cs.LG) Computer Science - Learning Computer Science - Robotics Range (mathematics) 020901 industrial engineering & automation Artificial Intelligence (cs.AI) 0202 electrical engineering electronic engineering information engineering Robot 020201 artificial intelligence & image processing Computer vision Artificial intelligence business Robotics (cs.RO) |
Zdroj: | ICRA |
DOI: | 10.48550/arxiv.1709.07857 |
Popis: | Instrumenting and collecting annotated visual grasping datasets to train modern machine learning algorithms can be extremely time-consuming and expensive. An appealing alternative is to use off-the-shelf simulators to render synthetic data for which ground-truth annotations are generated automatically. Unfortunately, models trained purely on simulated data often fail to generalize to the real world. We study how randomized simulated environments and domain adaptation methods can be extended to train a grasping system to grasp novel objects from raw monocular RGB images. We extensively evaluate our approaches with a total of more than 25,000 physical test grasps, studying a range of simulation conditions and domain adaptation methods, including a novel extension of pixel-level domain adaptation that we term the GraspGAN. We show that, by using synthetic data and domain adaptation, we are able to reduce the number of real-world samples needed to achieve a given level of performance by up to 50 times, using only randomly generated simulated objects. We also show that by using only unlabeled real-world data and our GraspGAN methodology, we obtain real-world grasping performance without any real-world labels that is similar to that achieved with 939,777 labeled real-world samples. Comment: 9 pages, 5 figures, 3 tables |
Databáze: | OpenAIRE |
Externí odkaz: |