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 1812.00568cs.ROcs.AIcs.CVcs.LG
收藏

视觉反馈:基于模型的深度强化学习用于基于视觉的机器人控制

Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Control

Georgia Institute of Technology(佐治亚理工学院);Twentieth Century Fox(二十世纪福克斯公司);Starfleet Academy(星际舰队学院)

Frederik Ebert, Chelsea Finn, Sudeep Dasari, Annie Xie, Alex Lee, Sergey Levine

摘要:

深度强化学习(RL)算法可以从原始的感官输入中学习复杂的机器人技能,但尚未实现深度学习方法在监督领域中所展示的广泛推广和适用性。我们提出了一种深度RL方法,该方法适用于现实世界的机器人任务,例如机器人操作,并有效地推广到从未见过的任务和对象。在这些环境中,地面真实奖励信号通常不可用,因此我们提出了一种基于自监督模型的方法,其中预测模型学习直接从原始感官读数(如相机图像)预测未来。在测试时,我们探索三种不同的目标规范方法:指定像素,其中用户指定所需的对象操作任务,通过选择特定的像素在图像和相应的目标位置,目标图像,其中所需的目标状态指定的图像,和图像分类器,定义空间的目标状态。我们的深度预测模型是使用机器人与数百个对象交互而自主和连续收集的数据进行训练的,无需人工监督。我们证明了视觉MPC可以推广到以前从未见过的对象-刚性和可变形-并使用相同的模型解决一系列用户定义的对象操作任务。

英文摘要:

Deep reinforcement learning (RL) algorithms can learn complex robotic skills from raw sensory inputs, but have yet to achieve the kind of broad generalization and applicability demonstrated by deep learning methods in supervised domains. We present a deep RL method that is practical for real-world robotics tasks, such as robotic manipulation, and generalizes effectively to never-before-seen tasks and objects. In these settings, ground truth reward signals are typically unavailable, and we therefore propose a self-supervised model-based approach, where a predictive model learns to directly predict the future from raw sensory readings, such as camera images. At test time, we explore three distinct goal specification methods: designated pixels, where a user specifies desired object manipulation tasks by selecting particular pixels in an image and corresponding goal positions, goal images, where the desired goal state is specified with an image, and image classifiers, which define spaces of goal states. Our deep predictive models are trained using data collected autonomously and continuously by a robot interacting with hundreds of objects, without human supervision. We demonstrate that visual MPC can generalize to never-before-seen objects---both rigid and deformable---and solve a range of user-defined object manipulation tasks using the same model.

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。

↑