arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2203.02230cs.LGcs.RO
收藏

用于模拟到真实深度强化学习的云边缘训练架构

Cloud-Edge Training Architecture for Sim-to-Real Deep Reinforcement Learning

Technical University of Munich(慕尼黑工业大学)

Hongpeng Cao, Mirco Theile, Federico G. Wyrwal, Marco Caccamo

摘要:

深度强化学习(Deep Reinforcement Learning,DRL)是一种通过与环境的交互来学习策略,从而解决复杂控制任务的有效方法。然而,DRL策略的培训需要大量的培训经验,这使得直接在物理系统上学习策略是不切实际的。模拟到真实的方法利用模拟来预训练DRL策略,然后将其部署到现实世界中。不幸的是,预训练策略的直接现实世界部署通常会由于不同的动态而遭受性能恶化,称为现实差距。最近的模拟到真实的方法,如域随机化和领域自适应,专注于提高预先训练的代理的鲁棒性。然而,模拟训练的策略通常需要使用真实世界的数据进行调整以达到最佳性能,由于真实世界样本的高成本,这是具有挑战性的。这项工作提出了一个分布式的云边缘架构,训练DRL代理在现实世界中的实时。在该架构中,推理和训练被分配给边缘和云,将实时控制回路与计算昂贵的训练回路分离。为了克服现实的差距,我们的架构利用仿真到现实迁移策略,继续模拟预训练代理的物理系统上的培训。我们证明了它的物理倒立摆控制系统的适用性,分析关键参数。真实世界的实验表明,我们的架构可以适应预先训练的DRL代理看不见的动态一致和有效的。

英文摘要:

Deep reinforcement learning (DRL) is a promising approach to solve complex control tasks by learning policies through interactions with the environment. However, the training of DRL policies requires large amounts of training experiences, making it impractical to learn the policy directly on physical systems. Sim-to-real approaches leverage simulations to pretrain DRL policies and then deploy them in the real world. Unfortunately, the direct real-world deployment of pretrained policies usually suffers from performance deterioration due to the different dynamics, known as the reality gap. Recent sim-to-real methods, such as domain randomization and domain adaptation, focus on improving the robustness of the pretrained agents. Nevertheless, the simulation-trained policies often need to be tuned with real-world data to reach optimal performance, which is challenging due to the high cost of real-world samples. This work proposes a distributed cloud-edge architecture to train DRL agents in the real world in real-time. In the architecture, the inference and training are assigned to the edge and cloud, separating the real-time control loop from the computationally expensive training loop. To overcome the reality gap, our architecture exploits sim-to-real transfer strategies to continue the training of simulation-pretrained agents on a physical system. We demonstrate its applicability on a physical inverted-pendulum control system, analyzing critical parameters. The real-world experiments show that our architecture can adapt the pretrained DRL agents to unseen dynamics consistently and efficiently.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑