The Role of Tactile Sensing in Learning and Deploying Grasp Refinement Algorithms
The Role of Tactile Sensing in Learning and Deploying Grasp Refinement Algorithms
复制标题
DOI：
10.1109/iros47612.2022.9981915
复制
发表时间：
2021-09
期刊：
影响因子：
--
通讯作者：
A. Koenig;Zixi Liu;Lucas Janson;R. Howe
中科院分区：
文献类型：
--
作者：
A. Koenig;Zixi Liu;Lucas Janson;R. Howe
机器人手设计中一个长期存在的问题是触觉传感必须有多精确。本文使用模拟触觉信号和强化学习（RL）框架研究抓取系统的感知需求。我们的第一个实验研究需要丰富的触觉传感的奖励RL为基础的把握细化算法多指机器人手。我们系统地整合不同层次的触觉数据到奖励使用分析把握稳定性指标。我们发现，在奖励中结合接触位置，法线和力的信息，长方体的平均成功率为95.4%，圆柱体为93.1%，球体为62.3%，手腕位置误差在0到7厘米之间，旋转误差在0到14度之间。这种基于接触的奖励比非触觉的二元奖励基线高出42.9%。我们的后续实验表明，当使用支持奖励的奖励进行训练时，控制策略的状态向量中触觉信息的使用大大减少，对于状态中没有触觉感知的情况，性能最多仅略微下降6.6%。由于策略不需要在测试时访问奖励信号，因此我们的工作意味着在支持抓取的手上训练的模型可以部署到具有较小传感器套件的机器人手上，从而可能大幅降低成本。
A long-standing question in robot hand design is how accurate tactile sensing must be. This paper uses simulated tactile signals and the reinforcement learning (RL) framework to study the sensing needs in grasping systems. Our first experiment investigates the need for rich tactile sensing in the rewards of RL-based grasp refinement algorithms for multi-fingered robotic hands. We systematically integrate different levels of tactile data into the rewards using analytic grasp stability metrics. We find that combining information on contact positions, normals, and forces in the reward yields the highest average success rates of 95.4% for cuboids, 93.1% for cylinders, and 62.3% for spheres across wrist position errors between 0 and 7 centimeters and rotational errors between 0 and 14 degrees. This contact-based reward outperforms a non-tactile binary-reward baseline by 42.9%. Our follow-up experiment shows that when training with tactile-enabled rewards, the use of tactile information in the control policy's state vector is drastically reducible at only a slight performance decrease of at most 6.6% for no tactile sensing in the state. Since policies do not require access to the reward signal at test time, our work implies that models trained on tactile-enabled hands are deployable to robotic hands with a smaller sensor suite, potentially reducing cost dramatically.