检索结果-内蒙古大学图书馆

Leveraging large language models for comprehensive locomotion control in humanoid robots design

Biomimetic Intelligence & robotics 2024年第4期4卷 48-63页

作者： Shilong Sun Chiyao Li Zida Zhao Haodong Huang Wenfu Xu School of Mechanical Engineering and Automation Harbin Institute of TechnologyShenzhen 518055China Guangdong Provincial Key Laboratory of Inteligent Morphing Mechanisms and Adaptive Robotics Shenzhen 518055China Key University Laboratory of Mechanism&Machine Theory and Intelligent Unmanned Systems of Guangdong Shenzhen 518055China

This paper investigates the utilization of large language models(LLMs)for the comprehensive control of humanoid robot *** reinforcement learning(RL)approaches for robot locomotion are resource-intensive and rely heavily on manually designed reward *** address these challenges,we propose a method that employs LLMs as the primary designer to handle key aspects of locomotion control,such as trajectory planning,inverse kinematics solving,and reward function *** using user-provided prompts,LLMs generate and optimize code,reducing the need for manual *** approach was validated through simulations in Unity,demonstrating that LLMs can achieve human-level performance in humanoid robot *** results indicate that LLMs can simplify and enhance the development of advanced locomotion control systems for humanoid robots.

关键词： Humanoid robots Large language models Locomotion control Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Aspect Optimal Sliding Mode Controller for a Mecanum Wheeled Robot 11

Multi-Aspect Optimal Sliding Mode Controller for a Mecanum W...

引用

11th RSI International Conference on robotics and Mechatronics, ICRoM 2023

作者： Allahyari, Sohrab Rahmanei, Hamid Moosavian, S. Ali A. K. N. Toosi University of Technology Center of Excellence in Robotics and Control Advanced Robotics and Automated Systems Laboratory Dept. of Mechanical Engineering Tehran Iran

ISBN: (纸本)9798350308105

Four-wheel drive Mecanum robots have gained attention due to their ability to move in all directions, which allows them to work in tight and complex environments. For this reason, ensuring precise control of these robots along predefined trajectories is crucial. This work presents a comparative study on the effects of parametric uncertainties, external disturbances, measurement noises, and initial deviations on the trajectory-Tracking performance of a 4-wheel drive Mecanum robot. First, the kinematic and kinetic models of the Mecanum robot are introduced. Then, the integral sliding mode control is designed, and proof of stability is provided. In addition, we explore other control strategies, including PID, feedback linearization, sliding mode, and adaptive sliding mode control, that are designed for this robot. Furthermore, the particle swarm optimization algorithm is hired for the controllers with a cost function as the total sum of average error, maximum error, and final error. Optimization has been done on all controllers to compare the results fairly. To appraise the effectiveness of the proposed control optimization methods, simulations are carried out in MATLAB/Simulink. The results reveal that the proposed ISMC method performs best in both defined scenarios compared to other controllers. However, it must be noted that the ISMC controller exhibits limitations when facing significant uncertainties in the dynamic model. This study underscores the effectiveness of our proposed control method and provides valuable insights into the practical implementation controller for the Mecanum robot in real-world applications. © 2023 IEEE.

关键词： Three term control systems

来源：评论

学校读者我要写书评

暂无评论

A novel freemium code SAND(v1.0)for generation of randomly packed beds

引用

Particuology 2024年第12期95卷 198-211页

作者： Nikita Shadymov Viacheslav Papkov Dmitry Pashchenko Samara State Technical University Molodogvardeiskaya 244 StSamara443100Russia Department of Mechanical Engineering(Robotics) Guangdong Technion-Israel Institute of Technology241 Daxue RoadShantouGuangdong515063China Faculty of Mechanical Engineering Technion-Israel Institute of TechnologyTechnion CityHaifa3200003Israel Guangdong Provincial Key Laboratory of Materials and Technologies for Energy Conversion Guangdong Technion—Israel Institute of TechnologyGuangdong515063China

The main current approaches for generation of the packed bed models are based on rigid body dynamics(RBD)and Newton's laws(discrete element methods-DEM).This paper deals with the development and analysis of a novel code based on analytical geometry approach for the packed bed *** architecture and main algorithms of the novel code are described and *** parameters of the packed bed generated via the novel code are compared with experimental data and packed beds generated via Blender(RBD),Yade(DEM).The novel code demonstrates many advantages,such as good correlation with experimental data,no overlaps between pellets in the packed bed,and a low computational time for packed bed *** packed bed model can be directly exported *** *** advantages are also demonstrated and *** novel code is attached to this paper and can be freely used by engineers and scientists.

关键词： Packed bed Discrete element method Python-based code CAD software

来源：评论

学校读者我要写书评

暂无评论

How Collective Intelligence Emerges in a Crowd of People Through Learned Division of Labor: A Case Study

How Collective Intelligence Emerges in a Crowd of People Thr...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Wang, Dekun Zhang, Hongwei School of Mechanical Engineering and Automation Harbin Institute of Technology Guangdong Provincial Key Laboratory of Intelligent Morphing Mechanisms and Adaptive Robotics Guangdong Shenzhen518055 China

ISBN: (纸本)9781665410205

This paper investigates the factors fostering collective intelligence (CI) through a case study of *LinVi's Experiment, where over 2000 human players collectively controll an avatar car. By conducting theoretical analysis and replicating observed behaviors through numerical simulations, we demonstrate how self-organized division of labor (DOL) among individuals fosters the emergence of CI and identify two essential conditions fostering CI by formulating this problem into a stability problem of a Markov Jump Linear System (MJLS). These conditions, independent of external stimulus, emphasize the importance of both elite and common players in fostering CI. Additionally, we propose an index for emergence of CI and a distributed method for estimating joint actions, enabling individuals to learn their optimal social roles without global action information of the whole crowd. © 2024 IEEE.

关键词： Linear systems

来源：评论

学校读者我要写书评

暂无评论

Bearing-Only Formation Tracking Control With Time-Varying Leaders' Velocity without Inter-Agent Communication 43

Bearing-Only Formation Tracking Control With Time-Varying Le...

引用

43rd Chinese Control Conference, CCC 2024

作者： Huang, Yuxiao Chen, Tianxing Zhang, Hongwei Harbin Institute of Technology Guangdong Provincial Key Laboratory of Intelligent Morphing Mechanisms and Adaptive Robotics School of Mechanical Engineering and Automation Guangdong Shenzhen518055 China

ISBN: (纸本)9789887581581

This paper studies the bearing-only formation tracking problem of leader-follower multi-agent systems, where the leaders move at a time-varying velocity. Different from the existing formation tracking methods, the proposed control protocol can be implemented without inter-agent communication. We construct the control protocol via a variable structure approach to track the time-varying velocity and almost achieve the target formation. We also show that the velocity tracking error will converge to zero and the position tracking error can be restricted in an upper bounded set while the upper bound can be arbitrarily minimized by increasing the control gain. Simulation results are presented to illustrate the effectiveness of the control protocol. © 2024 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Target tracking

来源：评论

学校读者我要写书评

暂无评论

Additive manufacturing of high-quality tritium breeding pebbles using ceramic precursors

引用

Journal of the European Ceramic Society 2024年第7期44卷 4719-4730页

作者： Su, Fang Li, Yuzeng Xing, Huanhuan Liu, Yu Liu, Zhiyuan Liu, Changyong Jiang, Cho-Pei Chen, Zhangwei Additive Manufacturing Institute Shenzhen University Shenzhen518060 China Guangdong Key Laboratory of Electromagnetic Control and Intelligent Robotics Shenzhen University Shenzhen518060 China College of Mechanical & Electrical Engineering National Taipei University of Technology Taipei10608 Taiwan State Key Laboratory for Manufacturing Systems Engineering Xi'an Jiaotong University Xi'an710049 China

In this study, the fabrication of advanced lithium orthosilicate (Li4SiO4) ceramic tritium breeding pebbles through photopolymerization using ceramic precursors (i.e., SiO2 and Li2CO3) is reported for the first time. The pebbles exhibit excellent mechanical and physical properties. Through optimization of manufacturing parameters and a geometrical correction model to compensate printing deviations, the printing accuracy of the pebbles was enhanced. The resulting pebbles had a diameter of approximately 1 mm, a sphericity of 1.01, a uniformly distributed pore structure, a phase purity > 95%, and a satisfying relative density reaching 91.39%. The average crush load 25.7 ± 0.3 N, surpassed that achieved by previously reported methods. Furthermore, the thermal conductivity was enhanced compared to that of solid-phase synthesized Li4SiO4 powder, facilitating the tritium release process for fusion applications. This study may serve as a guidance for designing appropriate ceramic breeders, providing a cost-effective approach applicable to produce other high-precision ceramic components with complex compositions. © 2024 Elsevier Ltd

关键词： Lithium compounds

来源：评论

学校读者我要写书评

暂无评论

An Efficient Planner for Dual-Arm Massage Robots in Constrained State

An Efficient Planner for Dual-Arm Massage Robots in Constrai...

引用

2024 IEEE International Conference on robotics and Biomimetics, ROBIO 2024

作者： Xu, Yuan Wang, Wenxiao Zhou, Lechuan Guo, Weichao Sheng, Xinjun Shanghai Jiao Tong University State Key Laboratory of Mechanical System and Vibration School of Mechanical Engineering Shanghai200240 China Shanghai Jiao Tong University Meta Robotics Institute Shanghai200240 China

ISBN: (纸本)9781665481090

Dual-arm massage robots necessitate reaching the massage pose along an optimal collision-free path in constrained state, where traditional path planning methods often exhibit inefficiency. This paper proposes a sampling-based planning algorithm that integrates lazy evaluation, transition testing, and bidirectional Rapidly-exploring Random Tree to reduce computational load, enhance planning efficiency, and improve the planner's ability to effectively explore the state space, avoid local minima, and converge to high-quality solutions. Comparative experiments conducted on a simulation platform across multiple algorithms have demonstrated the efficacy of the proposed planner, achieving a 95 % planning success rate within 1 second. The paths generated by the algorithm exhibit superior quality and enhanced consistency. In summary, the proposed algorithm is capable of producing asymptotically optimal, collision-free paths for a dual-arm massage robot. © 2024 IEEE.

关键词： Robot programming

来源：评论

学校读者我要写书评

暂无评论

Crosstalk-Free Dual-Mode Tactile Skin Based on 3D-Constrained Electric Field and Isolated Porous Iontronic Dielectric

Crosstalk-Free Dual-Mode Tactile Skin Based on 3D-Constraine...

引用

2024 IEEE International Conference on robotics and Biomimetics, ROBIO 2024

作者： Zhao, Yanchao Guo, Weichao Sheng, Xinjun School of Mechanical Engineering Shanghai Jiao Tong University State Key Laboratory of Mechanical System and Vibration Shanghai200240 China Meta Robotics Institute Shanghai Jiao Tong University Shanghai200240 China

ISBN: (纸本)9781665481090

Proximity-pressure dual-mode capacitive sensor is a novel sensing form to simultaneously detect distance and pressure information through a single device. In recent years, much effort has been devoted to improve its performance. However, it still suffers from robustness issues such as crosstalk and noise in practical applications. Electromagnetic coupling crosstalk caused by the outward electric fields as well as the response from unexpected directions also limit its distributed or array arrangements. In this regard, this paper proposes 3D-shielding structure as well as isolated iontronic dielectric designs to reduce the effects of electromagnetic/mechanical crosstalk. Through simulations and experiments, we demonstrate the effectiveness of the proposed design and optimise the structural parameters. The finally obtained sensing skin has an excellent crosstalk rejection ratio (15.68 dB, approximately three times improvement) while maintaining high response sensitivity. © 2024 IEEE.

关键词： Crosstalk

来源：评论

学校读者我要写书评

暂无评论

Multi-DoF Continuous Estimation for Wrist Torques Using Convolutional Neural Network

Multi-DoF Continuous Estimation for Wrist Torques Using Conv...

引用

2024 IEEE International Conference on robotics and Biomimetics, ROBIO 2024

作者： Fang, Yun Yu, Yang Guo, Weichao Sheng, Xinjun School of Mechanical Engineering Shanghai Jiao Tong University State Key Laboratory of Mechanical System and Vibration Shanghai200240 China Meta Robotics Institute Shanghai Jiao Tong University Shanghai200240 China

ISBN: (纸本)9781665481090

Using surface electromyography (sEMG) signals for simultaneous and proportional control (SPC) is appealing for natural control in human-machine interaction (HMI). Introducing spatial information across sEMG channels in the decoding process provides a more complete approach than methods that focus only on time and frequency domain features. In this article, we proposed an sEMG-based model for thesi-multaneous estimation of two degree-of-freedom (DoF) wrist torques, namely, Pro-Sup and Fle-Ext. Specifically, the 8 × 24 channels sEMG signals were transformed into images, which further served as inputs in a 29-layer convolutional neural network to estimate multi-DoF wrist torques. To prove the superiority of the CNN-based model, comparisons with other regression methods were performed on eight healthy subjects and two amputees in terms of the determination coefficient (R2) and normalized root mean square error (nRMSE). The results demonstrated that the CNN-based model (CNN-WAF) outperformed two conventional regression methods on estimation accuracy (R2: 0.851 ± 0.024 for Pro-Sup and 0.863 ± 0.016 for Fle-Ext;nRMSE: 6.933% ± 0.872% for Pro-Sup and 6.830% ± 0.856% for Fle-Ext). The average filtering in data preprocessing greatly improves (p © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A High-Speed Centerline Extraction Method for Multiple Laser Stripe based on Hessian Matrix

A High-Speed Centerline Extraction Method for Multiple Laser...

引用

International Conference on Biomimetic Intelligence and robotics & Medical robotics Forum, ICBIR+MRF 2024

作者： Shi, Juntian Guo, Weichao Sheng, Xinjun State Key Laboratory of Mechanical System and Vibration School of Mechanical Engineering Shanghai Jiao Tong University Shanghai200240 China Meta Robotics Institute Shanghai Jiao Tong University Shanghai200240 China

Laser centerline extraction is a critical step in the machine vision-aided measurement process, especially in the field of aerospace drilling and riveting. We propose a novel and efficient algorithm for the extraction of multiple laser centerlines, which consists of three major parts: laser linewidth estimation, ROI (Region of Interest) extraction, and precise positioning. In the linewidth estimation part, a method based on second-order moment is used to estimate the linewidth, and the scanning frame to linewidth ratio is adopted as the convergence condition to achieve a stable estimation result. In the ROI extraction part, we use a scanning frame to perform raster scanning on the entire image, and calculate the feature function threshold based on the sampling and grayscale volume method to complete ROI extraction. In the precise positioning part, we employ an improved method based on the Hessian matrix to reduce computation time, and use a λ filtering method to eliminate noise. Finally, the K-means method is used to classify the center points belonging to different laser centerlines, and RANSAC is utilized to perform line fitting. The final results demonstrate that the proposed algorithm has certain improvements in accuracy and also exhibits high computational efficiency. © 2024 The Authors.

关键词： Linewidth

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：