检索结果-内蒙古大学图书馆

Rover-IRL: Inverse Reinforcement learning With Soft Value Iteration Networks for Planetary Rover Path Planning

IEEE robotics AND automation LETTERS 2019年第2期4卷 1387-1394页

作者： Pflueger, Max Agha, Ali Sukhatme, Gaurav S. Univ Southern Calif Dept Comp Sci Los Angeles CA 90089 USA CALTECH Jet Prop Lab 4800 Oak Grove Dr Pasadena CA 91109 USA

Planetary rovers, such as those currently on Mars, face difficult path planning problems, both before landing during the mission planning stages as well as once on the ground. In this work, we present a new approach to these planning problems based on inverse reinforcement learning using deep convolutional networks and value iteration networks (VIN) as important internal structures. VIN are an approximation of the value iteration (VI) algorithm implemented with convolutional neural networks to make VI fully differentiable. We propose a modification to the value iteration recurrence, referred to as the soft value iteration network (SVIN). SVIN is designed to produce more effective training gradients through the VIN. It relies on an internal soft policy model, where the policy is represented with a probability distribution over all possible actions, rather than a deterministic policy that returns only the best action. We demonstrate the effectiveness of our proposed architecture in both a grid world dataset as well as a highly realistic synthetic dataset generated from currently deployed rover mission planning tools and real Mars imagery.

关键词： Space robotics and automation deep learning in robotics and automation learning from demonstration

来源：评论

学校读者我要写书评

暂无评论

Collision Detection for Industrial Collaborative Robots: A deep learning Approach

引用

IEEE robotics AND automation LETTERS 2019年第2期4卷 740-746页

作者： Heo, Young Jin Kim, Dayeon Lee, Woongyong Kim, Hyoungkyun Park, Jonghoon Chung, Wan Kyun Neuromeka Seoul 06023 South Korea POSTECH Dept Mech Engn Pohang 790784 South Korea

With increased human-robot interactions in industrial settings, a safe and reliable collision detection framework has become an indispensable element of collaborative robots. The conventional framework detects collisions by estimating collision monitoring signals with a particular type of observer, which is followed by collision decision processes. This results in unavoidable tradeoff between sensitivity to collisions and robustness to false alarms. In this study, we propose a collision detection framework (CollisionNet) based on a deep learning approach. We designed a deep neural network model to learn robot collision signals and recognize any occurrence of a collision. This data-driven approach unifies feature extraction from high-dimensional signals and the decision processes. CollisionNet eliminates heuristic and cumbersome nature of the traditional decision processes, showing high detection performance and generalization capability in real time. We verified the performance of the proposed framework through various experiments.

关键词： deep learning in robotics and automation physical human-robot interaction

来源：评论

学校读者我要写书评

暂无评论

Monocular Camera Based Fruit Counting and Mapping With Semantic Data Association

引用

IEEE robotics AND automation LETTERS 2019年第3期4卷 2296-2303页

作者： Liu, Xu Chen, Steven W. Liu, Chenhao Shivakumar, Shreyas S. Das, Jnaneshwar Taylor, Camillo J. Underwood, James Kumar, Vijay Univ Penn Grasp Lab Philadelphia PA 19104 USA Arizona State Univ Sch Earth & Space Explorat Tempe AZ 85281 USA Univ Sydney Australian Ctr Field Robot Camperdown NSW 2006 Australia

In this letter, we present a cheap, lightweight, and fast fruit counting pipeline. Our pipeline relies only on a monocular camera, and achieves counting performance comparable to a state-of-the-art fruit counting system that utilizes an expensive sensor suite including a monocular camera, LiDAR and GPS/INS on amango dataset. Our pipeline begins with a fruit and tree trunk detection component that uses state-of-the-art convolutional neural networks (CNNs). It then tracks fruits and tree trunks across images, with a Kalman Filter fusing measurements from the CNN detectors and an optical flow estimator. Finally, fruit count and map are estimated by an efficient fruit-as-feature semantic structure from motion algorithm that converts two-dimensional (2-D) tracks of fruits and trunks into 3-D landmarks, and uses these landmarks to identify double counting scenarios. There are many benefits of developing such a low cost and lightweight fruit counting system, including applicability to agriculture in developing countries, where monetary constraints or unstructured environments necessitate cheaper hardware solutions.

关键词： robotics in agriculture and forestry deep learning in robotics and automation visual tracking mapping object detection segmentation and categorization

来源：评论

学校读者我要写书评

暂无评论

Sensor Transfer: learning Optimal Sensor Effect Image Augmentation for Sim-to-Real Domain Adaptation

引用

IEEE robotics AND automation LETTERS 2019年第3期4卷 2431-2438页

作者： Carlson, Alexandra Skinner, Katherine A. Vasudevan, Ram Johnson-Roberson, Matthew Univ Michigan Robot Inst Ann Arbor MI 48109 USA Univ Michigan Dept Mech Engn Ann Arbor MI 48109 USA Univ Michigan Dept Naval Architecture & Marine Engn Ann Arbor MI 48109 USA

Performance on benchmark datasets has drastically improved with advances in deep learning. Still, cross-dataset generalization performance remains relatively low due to the domain shift that can occur between two different datasets. This domain shift is especially exaggerated between synthetic and real datasets. Significant research has been done to reduce this gap, specifically via modeling variation in the spatial layout of a scene, such as occlusions, and scene environmental factors, such as time of day and weather effects. However, few works have addressed modeling the variation in the sensor domain as a means of reducing the synthetic to real domain gap. The camera or sensor used to capture a dataset introduces artifacts into the image data, which are unique to the sensor model, suggesting that sensor effects may also contribute to domain shift. To address this, we propose a learned augmentation network composed of physically-based augmentation functions. Our proposed augmentation pipeline transfers specific effects of the sensor model-chromatic aberration, blur, exposure, noise, and color temperature-from a real dataset to a synthetic dataset. We provide experiments which demonstrate that augmenting synthetic training datasets with the proposed learned augmentation framework reduces the domain gap between synthetic and real domains for object detection in urban driving scenes.

关键词： deep learning in robotics and automation Visual learning Computer Vision for Other Robotic Applications Simulation and Animation

来源：评论

学校读者我要写书评

暂无评论

From Pixels to Percepts: Highly Robust Edge Perception and Contour Following Using deep learning and an Optical Biomimetic Tactile Sensor

引用

IEEE robotics AND automation LETTERS 2019年第2期4卷 2101-2107页

作者： Lepora, Nathan E. Church, Alex de Kerckhove, Conrad Hadsell, Raia Lloyde, John Univ Bristol Dept Engn Math Bristol BS8 1QU Avon England Univ Bristol Bristol Robot Lab Bristol BS8 1QU Avon England Google DeepMind London N1C 4AG England

deep learning has the potential to have same the impact on robot touch as it has had on robot vision. Optical tactile sensors act as a bridge between the subjects by allowing techniques from vision to be applied to touch. In this letter, we apply deep learning to an optical biomimetic tactile sensor, the TacTip, which images an array of papillae (pins) inside its sensing surface analogous to structures within human skin. Our main result is that the application of a deep convolutional neural network can give reliable edge perception, and, thus a robust policy for planning contact points to move around object contours. Robustness is demonstrated over several irregular and compliant objects with both tapping and continuous sliding, using a model trained only by tapping onto a disk. These results relied on using techniques to encourage generalization to tasks beyond which the model was trained. We expect this is a generic problem in practical applications of tactile sensing that deep learning will solve.

关键词： Force and tactile sensing biomimetics deep learning in robotics and automation

来源：评论

学校读者我要写书评

暂无评论

deep Reinforcement learning Robot for Search and Rescue Applications: Exploration in Unknown Cluttered Environments

引用

IEEE robotics AND automation LETTERS 2019年第2期4卷 610-617页

作者： Niroui, Farzad Zhang, Kaicheng Kashino, Zendai Nejat, Goldie Univ Toronto Dept Mech & Ind Engn Autonomous Syst & Biomechatron Lab Toronto ON M5S 3G8 Canada

Rescue robots can be used in urban search and rescue (USAR) applications to perform the important task of exploring unknown cluttered environments. Due to the unpredictable nature of these environments, deep learning techniques can be used to perform these tasks. In this letter, we present the first use of deep learning to address the robot exploration task in USAR applications. In particular, we uniquely combine the traditional approach of frontier-based exploration with deep reinforcement learning to allow a robot to autonomously explore unknown cluttered environments. Experiments conducted with a mobile robot in unknown cluttered environments of varying sizes and layouts showed that the proposed exploration approach can effectively determine appropriate frontier locations to navigate to, while being robust to different environment layouts and sizes. Furthermore, a comparison study with other frontier exploration approaches showed that our learning-based frontier exploration technique was able to explore more of an environment earlier on, allowing for potential identification of a larger number of victims at the beginning of the time-critical exploration task.

关键词： Autonomous agents deep learning in robotics and automation search and rescue robots

来源：评论

学校读者我要写书评

暂无评论

Multi-Task Regression-Based learning for Autonomous Unmanned Aerial Vehicle Flight Control Within Unstructured Outdoor Environments

引用

IEEE robotics AND automation LETTERS 2019年第4期4卷 4116-4123页

作者： Maciel-Pearson, Bruna G. Akcay, Samet Atapour-Abarghouei, Amir Holder, Christopher Breckon, Toby P. Univ Durham Dept Comp Sci Durham DH1 3LE England

Increased growth in the global unmanned aerial vehicles (UAV) (drone) industry has expanded possibilities for fully autonomous UAV applications. A particular application which has in part motivated this research is the use of UAV in wide area search and surveillance operations in unstructured outdoor environments. The critical issue with such environments is the lack of structured features that could aid in autonomous flight, such as road lines or paths. In this letter, we propose an end-to-end multi-task regression-based learning approach capable of defining flight commands for navigation and exploration under the forest canopy, regardless of the presence of trails or additional sensors (i.e., GPS). Training and testing are performed using a software in the loop pipeline, which allows for a detailed evaluation against state-of-the-art pose estimation techniques. Our extensive experiments demonstrate that our approach excels in performing dense exploration within the required search perimeter, is capable of covering wider search regions, generalises to previously unseen and unexplored environments and outperforms contemporary state-of-the-art techniques.

关键词： Aerial systems: perception and autonomy autonomous vehicle navigation computer vision for other robotic applications deep learning in robotics and automation

来源：评论

学校读者我要写书评

暂无评论

Understanding Natural Language Instructions for Fetching Daily Objects Using GAN-Based Multimodal Target-Source Classification

引用

IEEE robotics AND automation LETTERS 2019年第4期4卷 3884-3891页

作者： Magassouba, Aly Sugiura, Komei Anh Trinh Quoc Kawai, Hisashi Natl Inst Informat & Commun Technol Kyoto 6190289 Japan

In this letter, we address multimodal language understanding with unconstrained fetching instruction for domestic service robots. A typical fetching instruction such as "Bring me the yellow toy from the white shelf" requires to infer the user intention, i.e., what object (target) to fetch and from where (source). To solve the task, we propose a multimodal target-source classifier model (MTCM), which predicts the region-wise likelihood of target and source candidates in the scene. Unlike other methods, MTCM can handle region-wise classification based on linguistic and visual features. We evaluated our approach that outperformed the state-of-the-art method on a standard dataset. We also extended MTCM with generative adversarial nets, and enabled simultaneous data augmentation and classification.

关键词： deep learning in robotics and automation domestic robots

来源：评论

学校读者我要写书评

暂无评论

Planning Approximate Exploration Trajectories for Model-Free Reinforcement learning in Contact-Rich Manipulation

引用

IEEE robotics AND automation LETTERS 2019年第4期4卷 4042-4047页

作者： Hoppe, Sabrina Lou, Zhongyu Hennes, Daniel Toussaint, Marc Bosch Corp Res D-71272 Stuttgart Germany Univ Stuttgart Machine Learning & Robot Lab D-70174 Stuttgart Germany

Recent progress in deep reinforcement learning has enabled simulated agents to learn complex behavior policies from scratch, but their data complexity often prohibits real-world applications. The learning process can be sped up by expert demonstrations but those can be costly to acquire. We demonstrate that it is possible to employ model-free deep reinforcement learning combined with planning to quickly generate informative data for a manipulation task. In particular, we use an approximate trajectory optimization approach for global exploration based on an upper confidence bound of the advantage function. The advantage is approximated by a network for Q-learning with separately updated streams for state value and advantage that allows ensembles to approximate model uncertainty for one stream only. We evaluate our method on new extensions to the classical peg-in-hole task, one of which is only solvable by active usage of contacts between peg tips and holes. The experimental evaluation suggests that our method explores more relevant areas of the environment and finds exemplar solutions faster-both on a real robot and in simulation. Combining our exploration with learning from demonstration outperforms state-of-the-art model-free reinforcement learning in terms of convergence speed for contact-rich manipulation tasks.

关键词： deep learning in robotics and automation dexterous manipulation learning and adaptive systems

来源：评论

学校读者我要写书评

暂无评论

Reinforcement learning based on movement primitives for contact tasks

引用

robotics AND COMPUTER-INTEGRATED MANUFACTURING 2020年 62卷 101863-000页

作者： Kim, Young-Loul Ahn, Kuk-Hyun Song, Jae-Bok Korea Univ Sch Mech Engn Seoul 02841 South Korea

Recently, robot learning through deep reinforcement learning has incorporated various robot tasks through deep neural networks, without using specific control or recognition algorithms. However, this learning method is difficult to apply to the contact tasks of a robot, due to the exertion of excessive force from the random search process of reinforcement learning. Therefore, when applying reinforcement learning to contact tasks, solving the contact problem using an existing force controller is necessary. A neural-network-based movement primitive (NNMP) that generates a continuous trajectory which can be transmitted to the force controller and learned through a deep deterministic policy gradient (DDPG) algorithm is proposed for this study. In addition, an imitation learning algorithm suitable for NNMP is proposed such that the trajectories similar to the demonstration trajectory are stably generated. The performance of the proposed algorithms was verified using a square peg-in-hole assembly task with a tolerance of 0.1 mm. The results confirm that the complicated assembly trajectory can be learned stably through NNMP by the proposed imitation learning algorithm, and that the assembly trajectory is improved by learning the proposed NNMP through the DDPG algorithm.

关键词： AI-based methods Force control deep learning in robotics and automation

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：