检索结果-内蒙古大学图书馆

What the Constant Velocity Model Can Teach Us About Pedestrian Motion Prediction

IEEE robotics AND automation LETTERS 2020年第2期5卷 1696-1703页

作者： Schoeller, Christoph Aravantinos, Vincent Lay, Florian Knoll, Alois Fortiss Res Inst Free State Bavaria D-80805 Munich Germany Tech Univ Munich D-80333 Munich Germany

Pedestrian motion prediction is a fundamental task for autonomous robots and vehicles to operate safely. In recent years many complex approaches based on neural networks have been proposed to address this problem. In this work we show that - surprisingly - a simple Constant Velocity Model can outperform even state-of-the-art neural models. This indicates that either neural networks are not able to make use of the additional information they are provided with, or that this information is not as relevant as commonly believed. Therefore, we analyze how neural networks process their input and how it impacts their predictions. Our analysis reveals pitfalls in training neural networks for pedestrian motion prediction and clarifies false assumptions about the problem itself. In particular, neural networks implicitly learn environmental priors that negatively impact their generalization capability, the motion history of pedestrians is irrelevant and interactions are too complex to predict. Our work shows how neural networks for pedestrian motion prediction can be thoroughly evaluated and our results indicate which research directions for neural motion prediction are promising in future.

关键词： Motion and path planning deep learning in robotics and automation

来源：评论

学校读者我要写书评

暂无评论

Visual Object Search by learning Spatial Context

引用

IEEE robotics AND automation LETTERS 2020年第2期5卷 1279-1286页

作者： Druon, Raphael Yoshiyasu, Yusuke Kanezaki, Asako Watt, Alassane Paul Sabatier Univ F-31330 Toulouse France CNRS AIST Joint Robot Lab Tsukuba Ibaraki 3058560 Japan Natl Inst Adv Ind Sci & Technol Tokyo 1350064 Japan Cent Supelec Rennes France

We present a visual navigation approach that uses context information to navigate an agent to find and reach a target object. To learn context from the objects present in the scene, we transform visual information into an intermediate representation called context grid which essentially represents how much the object at the location is semantically similar to the target object. As this representation can encode the target object and other objects together, it allows us to navigate an agent in a human-inspired way: the agent will go to the likely place by seeing surrounding context objects in the beginning when the target is not visible and, once the target object comes into sight, it will reach the target quickly. Since context grid does not directly contain visual or semantic feature values that change according to introductions of new objects, such as new instances of the same object with different appearance or an object from a slightly different class, our navigation model generalizes well to unseen scenes/objects. Experimental results show that our approach outperforms previous approaches in navigating in unseen scenes, especially for broad scenes. We also evaluated human performances in the target-driven navigation task and compared with machine learning based navigation approaches including this work.

关键词： deep learning in robotics and automation visual-based navigation autonomous agents

来源：评论

学校读者我要写书评

暂无评论

deepFactors: Real-Time Probabilistic Dense Monocular SLAM

引用

IEEE robotics AND automation LETTERS 2020年第2期5卷 721-728页

作者： Czarnowski, Jan Laidlow, Tristan Clark, Ronald Davison, Andrew J. Imperial Coll London Dyson Robot Lab London SW7 2AZ England

The ability to estimate rich geometry and camera motion from monocular imagery is fundamental to future interactive robotics and augmented reality applications. Different approaches have been proposed that vary in scene geometry representation (sparse landmarks, dense maps), the consistency metric used for optimising the multi-view problem, and the use of learned priors. We present a SLAM system that unifies these methods in a probabilistic framework while still maintaining real-time performance. This is achieved through the use of a learned compact depth map representation and reformulating three different types of errors: photometric, reprojection and geometric, which we make use of within standard factor graph software. We evaluate our system on trajectory estimation and depth reconstruction on real-world sequences and present various examples of estimated dense geometry.

关键词： deep learning in robotics and automation mapping SLAM

来源：评论

学校读者我要写书评

暂无评论

learning Matchable Image Transformations for Long-Term Metric Visual Localization

引用

IEEE robotics AND automation LETTERS 2020年第2期5卷 1492-1499页

作者： Clement, Lee Gridseth, Mona Tomasi, Justin Kelly, Jonathan Univ Toronto Inst Aerosp Studies UTIAS Space & Terr Autonomous Robot Syst STARS Lab Toronto ON M3H 5T6 Canada UTIAS ASRL N York ON M3H 5T6 Canada

Long-term metric self-localization is an essential capability of autonomous mobile robots, but remains challenging for vision-based systems due to appearance changes caused by lighting, weather, or seasonal variations. While experience-based mapping has proven to be an effective technique for bridging the 'appearance gap,' the number of experiences required for reliable metric localization over days or months can be very large, and methods for reducing the necessary number of experiences are needed for this approach to scale. Taking inspiration from color constancy theory, we learn a nonlinear RGB-to-grayscale mapping that explicitly maximizes the number of inlier feature matches for images captured under different lighting and weather conditions, and use it as a pre-processing step in a conventional single-experience localization pipeline to improve its robustness to appearance change. We train this mapping by approximating the target non-differentiable localization pipeline with a deep neural network, and find that incorporating a learned low-dimensional context feature can further improve cross-appearance feature matching. Using synthetic and real-world datasets, we demonstrate substantial improvements in localization performance across day-night cycles, enabling continuous metric localization over a 30-hour period using a single mapping experience, and allowing experience-based localization to scale to long deployments with dramatically reduced data requirements.

关键词： deep learning in robotics and automation visual learning visual-based navigation localization

来源：评论

学校读者我要写书评

暂无评论

Robotic Understanding of Spatial Relationships Using Neural-Logic learning

Robotic Understanding of Spatial Relationships Using Neural-...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Yan, Fujian Wang, Dali He, Hongsheng Wichita State Univ Dept Elect Engn & Comp Sci Wichita KS 67260 USA Oak Ridge Natl Lab ORNL Artificial Intelligence AI Team Oak Ridge TN USA

ISBN: (纸本)9781728162126

Understanding spatial relations of objects is critical in many robotic applications such as grasping, manipulation, and obstacle avoidance. Humans can simply reason object's spatial relations from a glimpse of a scene based on prior knowledge of spatial constraints. The proposed method enables a robot to comprehend spatial relationships among objects from RGB-D data. This paper proposed a neural-logic learning framework to learn and reason spatial relations from raw data by following logic rules on spatial constraints. The neural-logic network consists of three blocks: grounding block, spatial logic block, and inference block. The grounding block extracts high-level features from the raw sensory data. The spatial logic blocks can predicate fundamental spatial relations by training a neural network with spatial constraints. The inference block can infer complex spatial relations based on the predicated fundamental spatial relations. Simulations and robotic experiments evaluated the performance of the proposed method.

关键词： spatial constraints neural-logic learning logic rules cognitive human-robot interaction deep learning in robotics and automation

来源：评论

学校读者我要写书评

暂无评论

learning to Generate 6-DoF Grasp Poses with Reachability Awareness

Learning to Generate 6-DoF Grasp Poses with Reachability Awa...

引用

IEEE International Conference on robotics and automation (ICRA)

作者： Lou, Xibai Yang, Yang Choi, Changhyun Univ Minnesota Dept Elect & Comp Engn Minneapolis MN 55455 USA Univ Minnesota Dept Comp Sci & Engn Minneapolis MN USA

ISBN: (纸本)9781728173955

Motivated by the stringent requirements of unstructured real-world where a plethora of unknown objects reside in arbitrary locations of the surface, we propose a voxel-based deep 3D Convolutional Neural Network (3D CNN) that generates feasible 6-DoF grasp poses in unrestricted workspace with reachability awareness. Unlike the majority of works that predict if a proposed grasp pose within the restricted workspace will be successful solely based on grasp pose stability, our approach further learns a reachability predictor that evaluates if the grasp pose is reachable or not from robot's own experience. To avoid the laborious real training data collection, we exploit the power of simulation to train our networks on a large-scale synthetic dataset. This work is an early attempt that simultaneously learns grasping reachability while proposing feasible grasp poses with 3D CNN. Experimental results in both simulation and real-world demonstrate that our approach outperforms several other methods and achieves 82.5% grasping success rate on unknown objects.

关键词： Grasping deep learning in robotics and automation Perception for Grasping and Manipulation

来源：评论

学校读者我要写书评

暂无评论

Robot Motion Planning in Learned Latent Spaces

引用

IEEE robotics AND automation LETTERS 2019年第3期4卷 2407-2414页

作者： Ichter, Brian Pavone, Marco Google Brain Mountain View CA 94043 USA Stanford Univ Dept Aeronaut & Astronaut Stanford CA 94305 USA

This letter presents latent sampling-based motion planning (L-SBMP), a methodology toward computing motion plans for complex robotic systems by learning a plannable latent representation. Recent works in control of robotic systems have effectively leveraged local, low-dimensional embeddings of high-dimensional dynamics. In this letter, we combine these recent advances with techniques from sampling-based motion planning (SBMP) in order to design a methodology capable of planning for high-dimensional robotic systems beyond the reach of traditional approaches (e.g., humanoids, or even systems where planning occurs in the visual space). Specifically, the learned latent space is constructed through an autoencoding network, a dynamics network, and a collision checking network, which mirror the three main algorithmic primitives of SBMP, namely state sampling, local steering, and collision checking. Notably, these networks can be trained through only raw data of the system's states and actions along with a supervising collision checker. Building upon these networks, an RRT-based algorithm is used to plan motions directly in the latent space-we refer to this exploration algorithm as learned latent RRT. This algorithm globally explores the latent space and is capable of generalizing to new environments. The overall methodology is demonstrated on two planning problems, namely a visual planning problem, whereby planning happens in the visual (pixel) space, and a humanoid robot planning problem.

关键词： Motion and path planning deep learning in robotics and automation

来源：评论

学校读者我要写书评

暂无评论

Semi-Supervised Gait Generation With Two Microfluidic Soft Sensors

引用

IEEE robotics AND automation LETTERS 2019年第3期4卷 2501-2507页

作者： Kim, Dooyoung Kim, Min Kwon, Junghan Park, Yong-Lae Jo, Sungho Korea Adv Inst Sci & Technol Sch Comp Daejeon 305701 South Korea Seoul Natl Univ Dept Mech Engn & Aerosp Engn Seoul 08826 South Korea Seoul Natl Univ Inst Adv Machines & Design Seoul 08826 South Korea

Nowadays, the use of deep learning for the calibration of soft wearable sensors has addressed the typical drawbacks of the microfluidic soft sensors, such as hysteresis and nonlinearity. However, previous studies have not yet resolved some of the design constraints such as the sensors are needed to he attached to the joints and many sensors are needed to track the human motion. Moreover, the previous methods also demand an excessive amount of data for sensor calibration which make the system impractical. In this letter, we present a gait motion generating method using only two microfluidic sensors. We select appropriate sensor positions with consideration of the deformation patterns of the lower-limb skins and mutual interference with soft actuators. Moreover, a semi-supervised deep learning model is proposed to reduce the size of calibration data. We evaluated the performance of the proposed model with various walking speeds. From the experiment, the proposed method showed a higher performance with smaller calibration dataset comparing to the other methods that are based on the supervised deep learning.

关键词： Wearable robots soft robot applications deep learning in robotics and automation

来源：评论

学校读者我要写书评

暂无评论

learning to Predict Ego-Vehicle Poses for Sampling-Based Nonholonomic Motion Planning

引用

IEEE robotics AND automation LETTERS 2019年第2期4卷 1053-1060页

作者： Banzhaf, Holger Sanzenbacher, Paul Baumann, Ulrich Zoellner, J. Marius Robert Bosch GmbH Corp Res Automated Driving D-71272 Renningen Germany FZI Res Ctr Informat Technol D-76131 Karlsruhe Germany

Sampling-based motion planning is an effective tool to compute safe trajectories for automated vehicles in complex environments. However, a fast convergence to the optimal solution can only be ensured with the use of problem-specific sampling distributions. Due to the large variety of driving situations within the context of automated driving, it is very challenging to manually design such distributions. This letter introduces, therefore, a data-driven approach utilizing a deep convolutional neural network (CNN): Given the current driving situation, future ego-vehicle poses can be directly generated from the output of the CNN allowing to guide the motion planner efficiently toward the optimal solution. A benchmark highlights that the CNN predicts future vehicle poses with a higher accuracy compared to uniform sampling and a state-of-the-art A*-based approach. Combining this CNN-guided sampling with the motion planner bidirectional RRT* reduces the computation time by up to an order of magnitude and yields a faster convergence to a lower cost as well as a success rate of 100% in the tested scenarios.

关键词： deep learning in robotics and automation non-holonomic motion planning motion and path planning

来源：评论

学校读者我要写书评

暂无评论

Normalization in Training U-Net for 2-D Biomedical Semantic Segmentation

引用

IEEE robotics AND automation LETTERS 2019年第2期4卷 1792-1799页

作者： Zhou, Xiao-Yun Yang, Guang-Zhong Imperial Coll London Hamlyn Ctr Robot Surg London SW7 2AZ England

Two-dimensional (2-D) biomedical semantic segmentation is important for robotic vision in surgery. Segmentation methods based on deep convolutional neural network (DCNN) can out-perform conventional methods in terms of both accuracy and levels of automation. One common issue in training a DCNN for biomedical semantic segmentation is the internal covariate shift where the training of convolutional kernels is encumbered by the distribution change of input features, hence both the training speed and performance are decreased. Batch normalization (BN) is the first proposed method for addressing internal covariate shift and is widely used. Instance normalization (IN) and layer normalization (LN) have also been proposed. Group normalization (GN) is proposed more recently and has not yet been applied to 2-D biomedical semantic segmentation (GN was used in 3-D biomedical semantic segmentation in [P.-Y. Kao, T. Ngo, A. Zhang, J. Chen, and B. Manjunath, Brain tumor segmentation and tractographic feature extraction from structural MR images for overall survival prediction 2018, arXiv:1807.07716], however, no specific validations on GN were given). Most DCNNs for biomedical semantic segmentation adopt BN as the normalization method by default, without reviewing its performance. In this letter, four normalization methods-BN, IN, LN, and GN are compared in details, specifically for 2-D biomedical semantic segmentation. U-Net is adopted as the basic DCNN structure. Three datasets regarding the right ventricle, aorta, and left ventricle are used for the validation. The results show that detailed subdivision of the feature map, i.e., GN with a large group number or IN, achieves higher accuracy. This accuracy improvement mainly comes from better model generalization. Codes are uploaded and maintained at Xiao-Yun Zhou's Github.

关键词： Computer vision formedical robotics deep learning in robotics and automation surgical robotics: planning visualbased navigation

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：