تقرير
BiGym: A Demo-Driven Mobile Bi-Manual Manipulation Benchmark
| العنوان: | BiGym: A Demo-Driven Mobile Bi-Manual Manipulation Benchmark |
|---|---|
| المؤلفون: | Chernyadev, Nikita, Backshall, Nicholas, Ma, Xiao, Lu, Yunfan, Seo, Younggyo, James, Stephen |
| سنة النشر: | 2024 |
| المجموعة: | Computer Science |
| مصطلحات موضوعية: | Computer Science - Robotics, Computer Science - Artificial Intelligence, Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning |
| الوصف: | We introduce BiGym, a new benchmark and learning environment for mobile bi-manual demo-driven robotic manipulation. BiGym features 40 diverse tasks set in home environments, ranging from simple target reaching to complex kitchen cleaning. To capture the real-world performance accurately, we provide human-collected demonstrations for each task, reflecting the diverse modalities found in real-world robot trajectories. BiGym supports a variety of observations, including proprioceptive data and visual inputs such as RGB, and depth from 3 camera views. To validate the usability of BiGym, we thoroughly benchmark the state-of-the-art imitation learning algorithms and demo-driven reinforcement learning algorithms within the environment and discuss the future opportunities. Comment: Project webpage: https://chernyadev.github.io/bigym/ |
| نوع الوثيقة: | Working Paper |
| URL الوصول: | http://arxiv.org/abs/2407.07788 |
| رقم الأكسشن: | edsarx.2407.07788 |
| قاعدة البيانات: | arXiv |
| الوصف غير متاح. |