Tactile Pose Estimation and Policy Learning for Unknown Object Manipulation
Tactile Pose Estimation and Policy Learning for Unknown Object Manipulation
复制标题
DOI：
10.48550/arxiv.2203.10685
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Tarik Kelestemur;Robert W. Platt;T. Padır
中科院分区：
文献类型：
--
作者：
Tarik Kelestemur;Robert W. Platt;T. Padır
物体姿态估计方法能够在非结构化环境中找到物体的位置。对于自主机器人操作来说，这是一项非常理想的技能，因为机器人需要估计物体的精确姿态才能对其进行操作。在本文中，我们研究了类别级物体的触觉姿态估计和操作问题。我们提出的方法使用了带有学习到的触觉观测模型和确定性运动模型的贝叶斯滤波器。随后，我们使用深度强化学习来训练策略，其中智能体使用来自贝叶斯滤波器的置信度估计。我们的模型在模拟环境中进行训练，并迁移到现实世界。我们通过一系列模拟和现实世界的实验分析了我们框架的可靠性和性能，并将我们的方法与基线工作进行了比较。我们的结果表明，学习到的触觉观测模型能够分别以2毫米和1度的分辨率对新物体的位置和方向进行姿态定位。此外，我们在一个开瓶任务上进行了实验，其中夹具需要达到期望的抓取状态。
Object pose estimation methods allow finding locations of objects in unstructured environments. This is a highly desired skill for autonomous robot manipulation as robots need to estimate the precise poses of the objects in order to manipulate them. In this paper, we investigate the problems of tactile pose estimation and manipulation for category-level objects. Our proposed method uses a Bayes filter with a learned tactile observation model and a deterministic motion model. Later, we train policies using deep reinforcement learning where the agents use the belief estimation from the Bayes filter. Our models are trained in simulation and transferred to the real world. We analyze the reliability and the performance of our framework through a series of simulated and real-world experiments and compare our method to the baseline work. Our results show that the learned tactile observation model can localize the pose of novel objects at 2-mm and 1-degree resolution for position and orientation, respectively. Furthermore, we experiment on a bottle opening task where the gripper needs to reach the desired grasp state.