检索结果-内蒙古大学图书馆

International Conference on Advanced robotics (ICAR)

作者： Joao P. Ferreira Manuel Crisostomo A. Paulo Coimbra Institute of Systems and Robotics Department of Electrical and Computer Engineering University of Coimbra Portugal

This paper describes the control of an autonomous biped robot capable of pulling a mass in the sagittal plane. The sagittal balance control uses an 8-link model of the robot, which is difficult to apply in real time due to the excessive computational effort. To solve this problem it is used a support vector regression (SVR) controller, which is able to perform the balance control in real time. As the ZMP balance control is non-linear, an SVR is appropriate. An SVR calculates the optimal hyper plane for the training data and computes faster than several other artificial intelligence techniques. An SVR that was trained based on experimental and simulation data, is used. This SVR uses the zero moment point (ZMP) position and its time variation as inputs, and the sagittal angle correction of the robot's body is obtained as the output. The ZMP is obtained making use of the force distribution on four force sensors placed under each robot's foot. This controller combines the use of the ankle and torso joints movements to correct the ZMP. The gait implemented in this biped is similar to a human gait that was acquired and scaled to the robot size. In this paper some experiments are also presented and the results show that the used gait combined with the SVR are appropriate to be used in the control of the biped robot, even when it drags a mass of about 65% of its own weight.

关键词： Weight control Robot sensing systems Force sensors Training data Artificial intelligence Computational modeling Foot Torso Joints Humans

来源：评论

学校读者我要写书评

暂无评论

Human silhouette volume reconstruction using a gravity-based virtual camera network

Human silhouette volume reconstruction using a gravity-based...

引用

International Conference on Information Fusion

作者： Hadi Aliakbarpour Jorge Dias Institute of Systems and Robotics Electrical and Computer Engineering Department University of Coimbra Portugal

The article represents a method to perform the Shape From Silhouette (SFS) of human, based on gravity sensing. A network of cameras is used to observe the scene. The extrinsic parameters among the cameras are initially unknown. An IMU is rigidly coupled to each camera in order to provide gravity and magnetic data. By applying a data fusion between each camera and its coupled IMU, it becomes possible to consider a downward-looking virtual camera for each camera within the network. Then extrinsic parameters among virtual cameras are estimated using the heights of two 3D points with respect to one camera within the network. Registered 2D points on the image plane of each camera is reprojected to its virtual camera image plane, using the concept of infinite homography. Such a virtual image plane is horizontal with a normal parallel to the gravity. The 2D points from the virtual image planes are back-projected onto the 3D space in order to make conic volumes of the observed object. From intersection of the created conic volumes from all cameras, the silhouette volume of the object is obtained. The experimental results validate both feasibility and effectiveness of the proposed method.

关键词： Cameras Three dimensional displays Calibration Image reconstruction Gravity Transmission line matrix methods Robot vision systems

来源：评论

学校读者我要写书评

暂无评论

A Human Gait Analyzer

A Human Gait Analyzer

引用

IEEE International Symposium on Intelligent Signal Processing

作者： Joao P. Ferreira Manuel Crisostomo A. Paulo Coimbra David Carnide Antonio Marto Institute of Systems and Robotics Department of Electrical and Computer Engineering University of Coimbra Portugal

ISBN: (纸本)9781424408290;1424408296

With this work, it is intended to do the tracking and analysis of the human motion, more specifically the gait. By using computational vision, it has been acquired the trajectories of defined control points in individuals' body, throughout time and space. These results are to be used afterwards in gait specification of biped robots. Several types of movement and the phases that compose a common system of capture and analysis of movement are referenced. Then, methods used in image processing and a description of existing gait types are detailed. Finally, the implemented software is presented and the results analyzed.

关键词： Humans Service robots Medical robotics computer vision Intelligent robots Biomedical imaging Kinematics Data mining Image processing Aerospace industry

来源：评论

学校读者我要写书评

暂无评论

AttDLNet: Attention-based DL Network for 3D LiDAR Place Recognition

arXiv

引用

arXiv 2021年

作者： Barros, Tiago Garrote, Luís Pereira, Ricardo Premebida, Cristiano Nunes, Urbano J. University of Coimbra Institute of Systems and Robotics Department of Electrical and Computer Engineering Portugal

LiDAR-based place recognition is one of the key components of SLAM and global localization in autonomous vehicles and robotics applications. With the success of DL approaches in learning useful information from 3D LiDARs, place recognition has also benefited from this modality, which has led to higher re-localization and loop-closure detection performance, particularly, in environments with significant changing conditions. Despite the progress in this field, the extraction of proper and efficient descriptors from 3D LiDAR data that are invariant to changing conditions and orientation is still an unsolved challenge. To address this problem, this work proposes a novel 3D LiDAR-based deep learning network (named AttDLNet) that uses a range-based proxy representation for point clouds and an attention network with stacked attention layers to selectively focus on long-range context and inter-feature relationships. The proposed network is trained and validated on the KITTI dataset and an ablation study is presented to assess the novel attention network. Results show that adding attention to the network improves performance, leading to efficient loop closures, and outperforming an established 3D LiDAR-based place recognition approach. From the ablation study, results indicate that the middle encoder layers have the highest mean performance, while deeper layers are more robust to orientation change. The code is publicly available at:https://***/Cybonic/AttDLNet. © 2021, CC BY.

关键词： Ablation

来源：评论

学校读者我要写书评

暂无评论

CNN-LIDAR pedestrian classification: combining range and reflectance data

CNN-LIDAR pedestrian classification: combining range and ref...

引用

IEEE International Conference on Vehicular Electronics and Safety (ICVES)

作者： Gledson Melotti Alireza Asvadi Cristiano Premebida Department of Electrical and Computer Engineering Institute of Systems and Robotics University of Coimbra Portugal

The use of multiple sensors in perception systems is becoming a consensus in the automotive and robotics industries. Camera is the most popular technology, however, radar and LIDAR are increasingly being adopted more often in protection and safety systems for object/obstacle detection. In this paper, we particularly explore the LIDAR sensor as an inter-modality technology which provides two types of data, range (distance) and reflectance (intensity return), and study the influence of high-resolution distance/depth (DM) and reflectance maps (RM) on pedestrian classification using a deep Convolutional Neural Network (CNN). Pedestrian protection is critical for advanced driver assistance system (ADAS) and autonomous driving, and it has regained particular attention recently for known reasons. In this work, CNN-LIDAR based pedestrian classification is studied in three distinct cases: (i) having a single modality as input in the CNN, (ii) by combining distance and reflectance measurements at the CNN input-level (early fusion), and (iii) combining outputs scores from two single-modal CNNs (late fusion). Distance and intensity (reflectance) raw data from LIDAR are transformed to high-resolution (dense) maps which allow a direct implementation on CNNs both as single or multi-channel inputs (early fusion approach). In terms of late-fusion, the outputs from individual CNNs are combined by means of non-learning rules, such as: minimum, maximum, average, product. Pedestrian classification is evaluated on a `binary classification' dataset created from the KITTI Vision Benchmark Suite, and results are shown for the three cases.

关键词： Laser radar Feature extraction Cameras Sensors Three-dimensional displays Safety Automotive engineering

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Boosting-SVM and SRM-SVM cascade classifiers in laser and vision-based pedestrian detection

Evaluation of Boosting-SVM and SRM-SVM cascade classifiers i...

引用

International Conference on Intelligent Transportation

作者： Oswaldo Ludwig Cristiano Premebida Urbano Nunes Rui Araújo Department of Electrical and Computer Engineering Institute of Systems and Robotics University of Coimbra Portugal

Pedestrian detection systems constitute an important field of research and development in computer vision, specially when applied in protection/safety systems in urban scenarios due to their direct impact in the society, specifically in terms of traffic casualties. In order to face such challenge, this work exploits some developments on statistical machine learning theory, in particular structural risk minimization (SRM) in a cascade ensemble. Namely, the ensemble applies the principle of SRM on a set of linear support vector machines (SVM). The linear SVM complexity, in the Vapnik sense, is controlled by choosing the dimension of the feature space in each cascade stage. To support experimental analysis, a multi-sensor dataset constituted by data from a LIDAR, a monocular camera, an IMU, encoder and a DGPS is introduced in this paper. The dataset, named Laser and Image Pedestrian Detection (LIPD) dataset, was collected in an urban environment, at day light conditions, using an electrical vehicle driven at low speed. Labeled pedestrians and non-pedestrians samples are also available for benchmarking purpose. The cascade of SVMs, trained with image-based features (HOG and COV descriptors), is used to detect pedestrian evidences on regions of interest (ROI) generated by a LIDAR-based processing system. Finally, the paper presents experimental results comparing the performance of a Boosting-SVM cascade and the proposed SRM-SVM cascade classifiers, in terms of detection errors.

关键词： Training Support vector machines Feature extraction Robot sensing systems Laser radar Detectors Complexity theory

来源：评论

学校读者我要写书评

暂无评论

Rejection of an external force in the sagittal plane applied on a biped robot using a neuro-fuzzy controller

Rejection of an external force in the sagittal plane applied...

引用

International Conference on Advanced robotics (ICAR)

作者： Joao P. Ferreira Manuel Crisostomo A. Paulo Coimbra Institute of Systems and Robotics Department of Electrical and Computer Engineering University of Coimbra Portugal

This paper describes the control of an autonomous biped robot capable to be subjected to external forces applied in the sagittal plane. A 1st order Takagi-Sugeno-Kang (TSK) type neuro fuzzy net, that was trained based on experimental and simulation data, is used. This net uses the zero moment point (ZMP) position and its variation as inputs, and the sagittal correction of the robot's body is obtained as the output. This control method is more accurate than the method based on an inverted pendulum model, which is based on a single mass concentration. The ZMP is obtained making use of the force distribution on four force sensors placed under each robot's foot. The gait implemented on our biped is similar to a human gait. A human gait was acquired and adapted to the robot size. In this paper some experiments are presented and the results show that the presented gait combined with the TSK neuro fuzzy net is appropriate to be used in the control of the biped robot, with and without an applied external force.

关键词： Force control Robot sensing systems Fuzzy neural networks Force sensors Humans Takagi-Sugeno-Kang model Weight control Biological system modeling Foot Fuzzy control

来源：评论

学校读者我要写书评

暂无评论

ZMP trajectory reference for the sagittal plane control of a biped robot based on a human CoP and gait

ZMP trajectory reference for the sagittal plane control of a...

引用

2009 IEEE/RSJ International Conference on Intelligent Robots and systems

作者： Manuel Crisóstomo A. Paulo Coimbra João Ferreira Institute of Systems and Robotics Department of Electrical and Computer Engineering University of Coimbra Portugal

This paper introduces two new important issues to be considered in the design of the zero moment point (ZMP) trajectory reference for the sagittal plane balance control of an autonomous walking biped robot with an human-like gait. ZMP trajectory reference generation is very important in the design and balance control of the walking of a biped robot. ZMP reference generation algorithms based on the linear inverted pendulum model (LIPM) and moving ZMP references in the swing phase have already been proposed with the ZMP trajectory during the swing phase being designed moving along a symmetric trajectory relative to the center of the foot. It was verified experimentally that in the human gait the ZMP trajectory moves along the foot in a way that it is shifted forward relative to its center. To take this into account a shift parameter is then proposed to move forward the X ZMP trajectory reference during the swing phase. It was also verified experimentally that in the human gait the ZMP trajectory amplitude depends on the swing time. Its variation law has been determined experimentally and it was verified that this range decreases as the swing time increases, reducing to zero for a static gait. It is then proposed a parameter H to take into account this variation with the swing time of the gait. Six experiments were carried out for three different X ZMP trajectory references. In order to evaluate and compare the performance of the biped robot using the three X ZMP trajectory references two performance indexes are proposed.

关键词： Humans Legged locomotion Intelligent robots Humanoid robots Robot sensing systems Control systems Foot Stability Kinematics Ground support

来源：评论

学校读者我要写书评

暂无评论

Multimodal Human Detection using RGB, Thermal and LiDAR modalities for Robotic Perception

Multimodal Human Detection using RGB, Thermal and LiDAR moda...

引用

IEEE International Conference on Automation Science and engineering (CASE)

作者： Kennedy O. S. Mota Luís Garrote Cristiano Premebida Department of Electrical and Computer Engineering Institute of Systems and Robotics University of Coimbra Portugal

ISBN: (数字)9798350358513

ISBN: (纸本)9798350358520

People detection is a relevant research topic in artificial perception with wide range of applications from security, surveillance, robotics to autonomous driving. Overcoming challenges in this field involves advanced algorithms, combination of machine learning approaches, as well as the use of sensory data e.g., from cameras and LiDARs. This work addresses the problem of people detection using YOLO, a state-of-the-art object detection method, trained on three distinct data sources LiDAR, RGB (color) and ‘thermal’ (long-wave infra-red) images. The rationale for combining multiple-sensory representation relies on the assumption that each sensor has its own advantages and disadvantages, but together they normally complement each other - specially in real-world conditions. LiDAR contributes to a physically-interpretable mapping of the environment, providing precise information regarding size/dimension and location of the objects, while RGB and thermal provide relevant textural features. The sensors have been calibrated w.r.t. each other thus, allowing the LiDAR’s point-clouds to be projected into the image plane, followed by an up-sampling step, to create dense-depth maps (DM) that enable direct use of the YOLO framework. To support the experiments, a new multi-sensory dataset has been collected using a mobile robot. Besides single-modality models, this paper also explores early and late-fusion strategies. Finally, the new dataset has been made available in a Github repository 1 .

关键词： YOLO Laser radar Machine learning algorithms Surveillance Soft sensors Thermal engineering Robot vision systems Security Mobile robots Software development management

来源：评论

学校读者我要写书评

暂无评论

Costmap-based Local Motion Planning using Deep Reinforcement Learning

Costmap-based Local Motion Planning using Deep Reinforcement...

引用

IEEE International Workshop on Robot and Human Communication (ROMAN)

作者： Luís Garrote João Perdiz Urbano J. Nunes Department of Electrical and Computer Engineering and the Institute of Systems and Robotics University of Coimbra Portugal

Local motion planning is an essential component of autonomous robot navigation systems as it involves generating collision-free trajectories for the robot in real-time, given its current position, the map of the environment and a goal. Considering an a priori goal path, computed by a global planner or as the output of a mission planning approach, this paper proposes a Two-Stream Deep Reinforcement Learning strategy for local motion planning that takes as inputs a local costmap representing the robot’s surrounding obstacles and a local costmap representing the nearest goal path. The proposed approach uses a Double Dueling Deep Q-Network and a new reward model to avoid obstacles while trying to maintain the lateral error between the robot and the goal path close to zero. Our approach enables the robot to navigate through complex environments, including cluttered spaces and narrow passages, while avoiding collisions with obstacles. Evaluation of the proposed approach was carried out in an in-house simulation environment, in five scenarios. Double and Double Dueling architectures were evaluated; the presented results show that the proposed strategy can correctly follow the desired goal path and, when needed, avoid obstacles ahead and recover back to following the goal path.

关键词：

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：