2013 IEEE International Conference on Robotics and Automation最新文献_第2页

Learning a dictionary of prototypical grasp-predicting parts from grasping experience 从抓取经验中学习典型抓取预测部分的字典

2013 IEEE International Conference on Robotics and Automation Pub Date : 2013-05-06 DOI: 10.1109/ICRA.2013.6630635

R. Detry, C. Ek, M. Madry, D. Kragic

引用次数: 120

Design of a direct-driven linear actuator for development of a cheetaroid robot 猎豹机器人直接驱动线性执行器的设计

2013 IEEE International Conference on Robotics and Automation Pub Date : 2013-05-06 DOI: 10.1109/ICRA.2013.6631144

Byeonghun Na, H. Choi, Kyoungchul Kong

引用次数: 7

Model-based imitation learning by probabilistic trajectory matching 基于模型的概率轨迹匹配模仿学习

2013 IEEE International Conference on Robotics and Automation Pub Date : 2013-05-06 DOI: 10.1109/ICRA.2013.6630832

Péter Englert, A. Paraschos, Jan Peters, M. Deisenroth

{"title":"Model-based imitation learning by probabilistic trajectory matching","authors":"Péter Englert, A. Paraschos, Jan Peters, M. Deisenroth","doi":"10.1109/ICRA.2013.6630832","DOIUrl":"https://doi.org/10.1109/ICRA.2013.6630832","url":null,"abstract":"One of the most elegant ways of teaching new skills to robots is to provide demonstrations of a task and let the robot imitate this behavior. Such imitation learning is a non-trivial task: Different anatomies of robot and teacher, and reduced robustness towards changes in the control task are two major difficulties in imitation learning. We present an imitation-learning approach to efficiently learn a task from expert demonstrations. Instead of finding policies indirectly, either via state-action mappings (behavioral cloning), or cost function learning (inverse reinforcement learning), our goal is to find policies directly such that predicted trajectories match observed ones. To achieve this aim, we model the trajectory of the teacher and the predicted robot trajectory by means of probability distributions. We match these distributions by minimizing their Kullback-Leibler divergence. In this paper, we propose to learn probabilistic forward models to compute a probability distribution over trajectories. We compare our approach to model-based reinforcement learning methods with hand-crafted cost functions. Finally, we evaluate our method with experiments on a real compliant robot.","PeriodicalId":259746,"journal":{"name":"2013 IEEE International Conference on Robotics and Automation","volume":"26 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120844626","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 51

Improving sparse roadmap spanners 改进稀疏路线图生成器

2013 IEEE International Conference on Robotics and Automation Pub Date : 2013-05-06 DOI: 10.1109/ICRA.2013.6631156

Andrew Dobson, Kostas E. Bekris

{"title":"Improving sparse roadmap spanners","authors":"Andrew Dobson, Kostas E. Bekris","doi":"10.1109/ICRA.2013.6631156","DOIUrl":"https://doi.org/10.1109/ICRA.2013.6631156","url":null,"abstract":"Roadmap spanners provide a way to acquire sparse data structures that efficiently answer motion planning queries with probabilistic completeness and asymptotic near-optimality. The current SPARS method provides these properties by building two graphs in parallel: a dense asymptotically-optimal roadmap based on PRM* and its spanner. This paper shows that it is possible to relax the conditions under which a sample is added to the spanner and provide guarantees, while not requiring the use of a dense graph. A key aspect of SPARS is that the probability of adding nodes to the roadmap goes to zero as iterations increase, which is maintained in the proposed extension. The paper describes the new algorithm, argues its theoretical properties and evaluates it against PRM* and the original SPARS algorithm. The experimental results show that the memory requirements of the method upon construction are dramatically reduced, while returning competitive quality paths with PRM*. There is a small sacrifice in the size of the final spanner relative to SPARS but the new method still returns graphs orders of magnitudes smaller than PRM*, leading to very efficient online query resolution.","PeriodicalId":259746,"journal":{"name":"2013 IEEE International Conference on Robotics and Automation","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121353243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 41

Learning task error models for manipulation 学习操作任务错误模型

2013 IEEE International Conference on Robotics and Automation Pub Date : 2013-05-06 DOI: 10.1109/ICRA.2013.6630935

P. Pastor, Mrinal Kalakrishnan, J. Binney, Jonathan Kelly, L. Righetti, G. Sukhatme, S. Schaal

{"title":"Learning task error models for manipulation","authors":"P. Pastor, Mrinal Kalakrishnan, J. Binney, Jonathan Kelly, L. Righetti, G. Sukhatme, S. Schaal","doi":"10.1109/ICRA.2013.6630935","DOIUrl":"https://doi.org/10.1109/ICRA.2013.6630935","url":null,"abstract":"Precise kinematic forward models are important for robots to successfully perform dexterous grasping and manipulation tasks, especially when visual servoing is rendered infeasible due to occlusions. A lot of research has been conducted to estimate geometric and non-geometric parameters of kinematic chains to minimize reconstruction errors. However, kinematic chains can include non-linearities, e.g. due to cable stretch and motor-side encoders, that result in significantly different errors for different parts of the state space. Previous work either does not consider such non-linearities or proposes to estimate non-geometric parameters of carefully engineered models that are robot specific. We propose a data-driven approach that learns task error models that account for such unmodeled non-linearities. We argue that in the context of grasping and manipulation, it is sufficient to achieve high accuracy in the task relevant state space. We identify this relevant state space using previously executed joint configurations and learn error corrections for those. Therefore, our system is developed to generate subsequent executions that are similar to previous ones. The experiments show that our method successfully captures the non-linearities in the head kinematic chain (due to a counterbalancing spring) and the arm kinematic chains (due to cable stretch) of the considered experimental platform, see Fig. 1. The feasibility of the presented error learning approach has also been evaluated in independent DARPA ARM-S testing contributing to successfully complete 67 out of 72 grasping and manipulation tasks.","PeriodicalId":259746,"journal":{"name":"2013 IEEE International Conference on Robotics and Automation","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127122391","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 31

Dynamic parameter identification of a 6 DOF industrial robot using power model 基于功率模型的六自由度工业机器人动态参数辨识

2013 IEEE International Conference on Robotics and Automation Pub Date : 2013-05-06 DOI: 10.1109/ICRA.2013.6630981

M. Gautier, S. Briot

引用次数: 24

Identification of geometrical and elastostatic parameters of heavy industrial robots 重型工业机器人几何及弹性静力参数辨识

2013 IEEE International Conference on Robotics and Automation Pub Date : 2013-05-06 DOI: 10.1109/ICRA.2013.6631098

A. Klimchik, Yier Wu, Claire Dumas-Lecerf, S. Caro, B. Furet, A. Pashkevich

引用次数: 45

Locating end-effector tips in automated micromanipulation 定位末端执行器尖端在自动微操作

2013 IEEE International Conference on Robotics and Automation Pub Date : 2013-05-06 DOI: 10.1109/ICRA.2013.6630803

Jun Liu, Zheng Gong, K. Tang, Zhe Lu, Yu Sun

{"title":"Locating end-effector tips in automated micromanipulation","authors":"Jun Liu, Zheng Gong, K. Tang, Zhe Lu, Yu Sun","doi":"10.1109/ICRA.2013.6630803","DOIUrl":"https://doi.org/10.1109/ICRA.2013.6630803","url":null,"abstract":"Locating end-effector tips is a prerequisite step in micromanipulation. The tip of micromanipulation tools is typically a few micrometers in size and highly delicate. In all existing automated micromanipulation systems, the process of locating the end-effector tip is conducted by a skilled operator, and the automation of this task has not been attempted. This paper presents a technique for automatically locating end-effector tips. The technique consists of programmed sweeping patterns, MHI (motion history image) end-effector detection, active contour for estimating end-effector positions, autofocusing and quad-tree search for locating end-effector tip, and finally visual servoing to position the tip to the center of the field of view. Two types of micromanipulation tools (micropipette representing single-ended tools and microgripper representing multi-ended tools) were used in experiments for testing. Quantitative results were reported in the speed and success rate of the auto-locating technique, based on over 500 trials. Furthermore, the effect of factors such as imaging mode and image processing parameter selections was also quantitatively discussed. Guidelines are provided for the implementation of the technique in order to achieve high efficiency and success rates.","PeriodicalId":259746,"journal":{"name":"2013 IEEE International Conference on Robotics and Automation","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125020133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 7

Learning a real time grasping strategy 学习实时抓取策略

2013 IEEE International Conference on Robotics and Automation Pub Date : 2013-05-06 DOI: 10.1109/ICRA.2013.6630634

Bidan Huang, S. El-Khoury, Miao Li, J. Bryson, A. Billard

引用次数: 47

Multi-sensor identity tracking with event graphs 带有事件图的多传感器身份跟踪

2013 IEEE International Conference on Robotics and Automation Pub Date : 2013-05-06 DOI: 10.1109/ICRA.2013.6631252

P. Morton, B. Douillard, J. Underwood

引用次数: 9