arXiv 2103.132672021-03-24cs.ROcs.LG

CLAMGen：基于多视图视觉的闭环手臂运动生成RL

CLAMGen: Closed-Loop Arm Motion Generation via Multi-view Vision-Based RL

Columbia University（哥伦比亚大学）

Iretiayo Akinola, Zizhao Wang, Peter Allen

摘要：

提出了一种基于视觉的强化学习（RL）方法，用于手臂到达问题的闭环轨迹生成。手臂轨迹生成是一个基本的机器人问题，它需要找到无碰撞路径来移动机器人的身体（例如手臂），以满足目标（例如将末端执行器放置在一个点上）。虽然经典方法通常需要环境模型来解决规划，搜索或优化问题，但基于学习的方法有望直接将观察映射到机器人动作。然而，由于各种原因，使用RL学习冲突避免策略仍然是一个挑战，包括但不限于部分可观察性、较差的探索、低样本效率和学习不稳定性。为了解决这些挑战，我们提出了一种剩余RL方法，该方法利用贪婪的目标达到RL策略作为基础来改善探索，并且该基础策略通过从图像中学习的剩余状态动作值和剩余动作来增强以避免障碍。此外，我们引入了新的学习目标和技术，以提高从多个图像视图的三维理解和我们的算法的样本效率。与RL基线相比，我们的方法在成功率方面取得了优异的性能。

英文摘要：

We propose a vision-based reinforcement learning (RL) approach for closed-loop trajectory generation in an arm reaching problem. Arm trajectory generation is a fundamental robotics problem which entails finding collision-free paths to move the robot's body (e.g. arm) in order to satisfy a goal (e.g. place end-effector at a point). While classical methods typically require the model of the environment to solve a planning, search or optimization problem, learning-based approaches hold the promise of directly mapping from observations to robot actions. However, learning a collision-avoidance policy using RL remains a challenge for various reasons, including, but not limited to, partial observability, poor exploration, low sample efficiency, and learning instabilities. To address these challenges, we present a residual-RL method that leverages a greedy goal-reaching RL policy as the base to improve exploration, and the base policy is augmented with residual state-action values and residual actions learned from images to avoid obstacles. Further more, we introduce novel learning objectives and techniques to improve 3D understanding from multiple image views and sample efficiency of our algorithm. Compared to RL baselines, our method achieves superior performance in terms of success rate.

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。

摘要：

英文摘要：

相关论文