检索结果-内蒙古大学图书馆

Human-robot Collaborative Minimum Time Search Through Sub-Priors in Ant Colony Optimization

IEEE robotICS AND AUTOMATION LETTERS 2024年第11期9卷 10216-10223页

作者： Gil, Oscar Sanfeliu, Alberto UPC CSIC Inst Robot & Informat Ind Barcelona 08028 Spain

Human-robot Collaboration (HRC) has evolved into a highly promising issue owing to the latest breakthroughs in Artificial Intelligence (AI) and Human-robot Interaction (HRI), among other reasons. This emerging growth increases the need to design multi-agent algorithms that can manage also human preferences. This letter presents an extension of the Ant Colony Optimization (ACO) meta-heuristic to solve the Minimum Time Search (MTS) task, in the case where humans and robots perform an object searching task together. The proposed model consists of two main blocks. The first one is a convolutional neural network (CNN) that provides the prior probabilities about where an object may be from a segmented image. The second one is the Sub-prior MTS-ACO algorithm (SP-MTS-ACO), which takes as inputs the prior probabilities and the particular search preferences of the agents in different sub-priors to generate search plans for all agents. The model has been tested in real experiments for the joint search of an object through a Vizanti web-based visualization in a tablet computer. The designed interface allows the communication between a human and our humanoid robot named IVO. The obtained results show an improvement in the search perception of the users without loss of efficiency.

关键词： robots Collaboration Search problems Computational modeling Probabilistic logic Optimization Metaheuristics Human-robot interaction robot sensing systems Planning Human-robot collaboration (HRC) path planning for multiple mobile robots or agents deep learning methods semantic scene understanding search and rescue robots

来源：评论

学校读者我要写书评

暂无评论

Cone Slalom With Automated Sports Car-;Trajectory Planning Algorithm

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2024年第1期73卷 362-374页

作者： Hanzalek, Zdenek Zahora, Jiri Sojka, Michal Czech Tech Univ Czech Inst Informat Robot & Cybernet Ind Informat Dept Prague 16000 Czech Republic

In this article, we present the system architecture and algorithms of an automated vehicle to perform a slalom. We demonstrate a novel trajectory planning algorithm based on optimization techniques using logic-based Benders decomposition, where an external loop optimizes the position of the nearest waypoint and an internal loop generates the optimal trajectory. The positions of the cones in this use case are unknown, but a mono camera and LiDAR detect them. They can be in a line or dispersed, have equal or unequal spacing, and the U-turns can be symmetric or asymmetric. A bicycle model is used to formulate a non-linear quadratic optimization problem aimed at optimal trajectory generation considering vehicle kinematics. Finally, the trajectory tracking control keeps the vehicle on the planned slalom trajectory while driving. The control system is interfaced with the vehicle via CAN and FlexRay buses. Much of the work was devoted to experiments with a real vehicle and fine-tuning the system parameters. During the validation of the system, interesting observations were made regarding the components' precision, frequency, and sensitivity.

关键词： Automobiles Trajectory Trajectory planning Optimization Planning Control systems Cameras Automated car slalom cone detection robotic operating system trajectory planning temporal properties

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Trajectory Segmentation and Gesture Recognition through Curvature Analysis and the Levenshtein Distance 7

Unsupervised Trajectory Segmentation and Gesture Recognition...

引用

7th Iberian robotics Conference

作者： Tapia, Guillem Colome, Adria Torras, Carme Inst RobOt & Informat Ind CSIC UPC Barcelona Spain

ISBN: (纸本)9798350376371;9798350376364

This work discusses segmentation and gesture recognition in human-driven robotic trajectories, a technique with applications in several sectors such as in robot-assisted minimally invasive surgery (RMIS) training. By decomposing entire movements -gestures- into smaller actions -sub-gestures-, we can address gesture recognition accurately. This paper extends a bottom-up approach used in surgical gesture segmentation and incorporates natural language processing (NLP) techniques to match sub-gestures with letters and treating gestures as words. We evaluated our algorithm using two different datasets with trajectories on 2D and 3D. This NLP-inspired model obtains an average F1-score of 94.25% in the segmentation tasks, an accuracy of 87.05% in the learning stage, and an overall accuracy of 88.79% in the fully automated execution. These results indicate that the method effectively identifies and interprets new surgical gestures autonomously without the need for human intervention.

关键词： Gesture recognition

来源：评论

学校读者我要写书评

暂无评论

3M-TRANSFORMER: A MULTI-STAGE MULTI-STREAM MULTIMODAL TRANSFORMER FOR EMBODIED TURN-TAKING PREDICTION 49

3M-TRANSFORMER: A MULTI-STAGE MULTI-STREAM MULTIMODAL TRANSF...

引用

49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Fatan, Mehdi Mincato, Emanuele Pintzou, Dimitra Dimiccoli, Mariella CSIC UPC Inst Robot & Informat Ind Barcelona Spain

ISBN: (纸本)9798350344868;9798350344851

Predicting turn-taking in multiparty conversations has many practical applications in human-computer/robot interaction. However, the complexity of human communication makes it a challenging task. Recent advances have shown that synchronous multi-perspective egocentric data can significantly improve turn-taking prediction compared to asynchronous, single-perspective transcriptions. Building on this research, we propose a new multimodal transformer-based architecture for predicting turn-taking in embodied, synchronized multi-perspective data. Our experimental results on the recently introduced EgoCom dataset show a substantial performance improvement of up to 14.01% on average compared to existing baselines and alternative transformer-based approaches. The source code, and the pre-trained models of our 3T-Transformer will be available upon acceptance.

关键词： cross-modal transformer turn-taking prediction embodied multi-perspective data audio-video-text.

来源：评论

学校读者我要写书评

暂无评论

Human Motion Trajectory Prediction Using the Social Force Model for Real-Time and Low Computational Cost Applications 6th

Human Motion Trajectory Prediction Using the Social Force Mo...

引用

6th Iberian robotics Conference (robot)

作者： Gil, Oscar Sanfeliu, Alberto CSIC UPC Inst Robot & Informat Ind Barcelona Spain

ISBN: (纸本)9783031586750;9783031586767

Human motion trajectory prediction is a very important functionality for human-robot collaboration, specifically in accompanying, guiding, or approaching tasks, but also in social robotics, self-driving vehicles, or security systems. In this paper, a novel trajectory prediction model, Social Force Generative Adversarial Network (SoFGAN), is proposed. SoFGAN uses a Generative Adversarial Network (GAN) and Social Force Model (SFM) to generate different plausible people trajectories reducing collisions in a scene. Furthermore, a Conditional Variational Autoencoder (CVAE) module is added to emphasize the destination learning. We show that our method is more accurate in making predictions in UCY or BIWI datasets than most of the current state-of-the-art models and also reduces collisions in comparison to other approaches. Through real-life experiments, we demonstrate that the model can be used in real-time without GPU's to perform good quality predictions with a low computational cost.

关键词： Human Motion Prediction Social Force Model Generative Adversarial Network Conditional Variational Autoencoder

来源：评论

学校读者我要写书评

暂无评论

Body Gesture Recognition to Control a Social Mobile robot 23

Body Gesture Recognition to Control a Social Mobile Robot

引用

18th Annual ACM/IEEE International Conference on Human-robot Interaction (HRI)

作者： Laplaza, Javier Romero, Ramon Sanfeliu, Alberto Garrell, Anais Inst Robot & Informat Ind Barcelona Barcelona Spain

ISBN: (纸本)9781450399708

In this work, we propose a gesture-based language to allow humans to interact with robots using their body in a natural way. We have created a new gesture detection model using neural networks and a new dataset of humans making a collection of body gestures to train this architecture. Furthermore, we compare body gesture communication with other communication channels to demonstrate the importance of adding this knowledge to robots. The presented approach is validated in diverse simulations and real-life experiments with non-trained volunteers. This attains promising results and establishes that it is a valuable framework for social robotic applications, such as human robot collaboration or human-robot interaction.

关键词： datasets neural networks HRI

来源：评论

学校读者我要写书评

暂无评论

Formulating the Unicycle on the Sphere Path Planning Problem as a Linear Time-Varying System

引用

IEEE TRANSACTIONS ON robotICS 2025年 41卷 3335-3347页

作者： Thomas, Federico Franch, Jaume UPC ETSEIB CSIC Inst Robot & Informat Ind Barcelona 08028 Spain UPC Dept Math Barcelona 08028 Spain

The kinematics, dynamics, and control of a unicycle moving without slipping on a plane has been extensively studied in the literature of nonholonomic mechanical systems. However, since planar motion can be seen as a limiting case of the motion on a sphere, we focus our analysis on the more general spherical case. This article introduces a novel approach to path planning for a unicycle rolling on a sphere while satisfying the nonslipping constraint. Our method is based on a simple yet effective idea: first, we model the system as a linear time-varying dynamic system. Then, leveraging the fact that certain such systems can be integrated under specific algebraic conditions, we derive a closed-form expression for the control variables. This formulation includes three free parameters, which can be tuned to generate a path connecting any two configurations of the unicycle. Notably, our approach requires no prior knowledge of nonholonomic system analysis, making it accessible to a broader audience.

关键词： robots Path planning Kinematics Finite element analysis Time-varying systems Standards Automobiles Angular velocity robot kinematics Quaternions Nonholonomic robots nonholonomic joints linear time-varying systems path planning

来源：评论

学校读者我要写书评

暂无评论

Voice Command Recognition for Explicit Intent Elicitation in Collaborative Object Transportation Tasks: a ROS-based Implementation 24

Voice Command Recognition for Explicit Intent Elicitation in...

引用

19th Annual ACM/IEEE International Conference on Human-robot Interaction (HRI)

作者： Dominguez-Vidal, J. E. Sanfeliu, Alberto UPC Inst Robot & Informat Ind CSIC Barcelona Spain

ISBN: (纸本)9798400703232

Voice command recognition remains relatively unexplored in robotics, with limited insight into user acceptance and real-world performance. In this work we try to address this by offering multiple voice command recognition models encapsulated in a single publicly available ROS node ready to be used by the robotics practitioner. We tested its actual performance with 10 volunteers of different nationalities whose first spoken language is not English. The obtained accuracy in these tests varies between 93.14% and 95.63% depending on the number of considered commands and model size. Finally, we conducted a user study with 23 new volunteers performing a human-robot collaborative transport task to test whether humans are willing to use this type of system despite having a non-negligible delay and failure rate. In addition to improvements in parameters such as comfort and trust in the robot, 86.9% of the volunteers chose this system over a technically more robust one.

关键词： Human-robot Interaction Intention Understanding Voice Recognition Human-centered Studies Natural Language Processing

来源：评论

学校读者我要写书评

暂无评论

PoseScript: Linking 3D Human Poses and Natural Language

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2025年第7期47卷 5146-5159页

作者： Delmas, Ginger Weinzaepfel, Philippe Lucas, Thomas Moreno-Noguer, Francesc Rogez, Gregory CSIC UPC Inst Robot & Informat Ind Barcelona 08028 Spain NAVER LABS Europe F-38240 Meylan France

Natural language plays a critical role in many computer vision applications, such as image captioning, visual question answering, and cross-modal retrieval, to provide fine-grained semantic information. Unfortunately, while human pose is key to human understanding, current 3D human pose datasets lack detailed language descriptions. To address this issue, we have introduced the PoseScript dataset. This dataset pairs more than six thousand 3D human poses from AMASS with rich human-annotated descriptions of the body parts and their spatial relationships. Additionally, to increase the size of the dataset to a scale that is compatible with data-hungry learning algorithms, we have proposed an elaborate captioning process that generates automatic synthetic descriptions in natural language from given 3D keypoints. This process extracts low-level pose information, known as "posecodes", using a set of simple but generic rules on the 3D keypoints. These posecodes are then combined into higher level textual descriptions using syntactic rules. With automatic annotations, the amount of available data significantly scales up (100k), making it possible to effectively pretrain deep models for finetuning on human captions. To showcase the potential of annotated poses, we present three multi-modal learning tasks that utilize the PoseScript dataset. Firstly, we develop a pipeline that maps 3D poses and textual descriptions into a joint embedding space, allowing for cross-modal retrieval of relevant poses from large-scale datasets. Secondly, we establish a baseline for a text-conditioned model generating 3D poses. Thirdly, we present a learned process for generating pose descriptions. These applications demonstrate the versatility and usefulness of annotated poses in various tasks and pave the way for future research in the field.

关键词： Three-dimensional displays Natural languages Task analysis Semantics Pipelines Legged locomotion Knee 3D human pose captioning description generation multi-modal learning natural language retrieval

来源：评论

学校读者我要写书评

暂无评论

Safety Methods for Cartesian Control of Redundant robotic Arms 7

Safety Methods for Cartesian Control of Redundant Robotic Ar...

引用

7th Iberian robotics Conference

作者： Antonio Delgado-Guerrero, Juan Colome, Adria Foix, Sergi Torras, Carme Inst RobOt & Informat Ind CSIC UPC Barcelona Spain

ISBN: (纸本)9798350376371;9798350376364

Over the last years, the interest in robot safety has increased, due to the improvements in applications such as assistive robotics, co-manipulation or manufacturing cobots. In these scenarios that include physical interactions between humans and robots, it is of upmost importance to have both passive and active safety layers that prevent robotic arms to behave in an unsafe manner. Ultimately, the safety measures that can be applied might depend on the sensors installed on a robot. robots with joint torque sensors often base their safety in reactive behaviors when a contact is detected. However, robots should be able to also have safety layers that do not depend on detected contacts. In this paper, we show several additions to robotic Cartesian controllers that can significantly improve safety in human-robot physical interaction for robots with joint position encoders only.

关键词： robotic arms

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：