检索结果-内蒙古大学图书馆

Autonomous search of real-life environments combining dynamical system-based path planning and unsupervised learning

引用

IEEE Access 2025年 13卷 101274-101296页

作者： Amadasun, Uyiosa Philip McNamee, Patrick Ahmadabadi, Zahra Nili Naseradinmousavi, Peiman San Diego State University Department of Mechanical Engineering San Diego 92182 CA United States Dynamic Systems and Intelligent Machines Lab United States Dynamic Systems and Control Lab United States

In recent years, advances have been made in chaotic coverage path planning (CCPP) for autonomous search and traversal of spaces with limited environmental cues. However, the field remains unfit for practical application due to limited experimental work addressing three critical challenges: (1) an obstacle avoidance technique that reduces halts or disruptions in continuous chaotic trajectories, (2) a means to spread chaotic trajectories across the environment (especially crucial for large and/or complex-shaped environments) that need to be covered, and (3) a real-time coverage calculation technique that is accurate and independent of cell size. This study develops a novel framework implemented in ROS to address these challenges. We compare our Chaotic Coverage Path Planning (CCPP) approach against Boustrophedon Coverage Path Planning (BCPP) in both simulated and real-world environments. Across all three simulated environments of increasing complexity, BCPP consistently achieved shorter coverage times than our CCPP algorithm. However, an interesting pattern emerged when analyzing the relative performance degradation. When comparing coverage times across environments, CCPP showed less performance degradation than BCPP. Specifically, when normalizing coverage times against the simplest environment, BCPP’s performance degraded by factors of 3.31 and 3.59 for the medium and most complex environments, respectively, while CCPP showed lower degradation factors of 2.57 and 2.35. This suggests that CCPP may be more adaptive to increasing environmental size and complexity, despite having longer absolute coverage times. © 2013 IEEE.

关键词： Autonomous robot Nonlinear dynamical system Path planning Unpredictable search

来源：评论

学校读者我要写书评

暂无评论

Gait Analysis with Trinocular Computer Vision Using Deep Learning

Gait Analysis with Trinocular Computer Vision Using Deep Lea...

引用

2023 IEEE International Conference on Image Processing Challenges and Workshops, ICIPCW 2023

作者： Stavrakakis, Odysseas Mastrogeorgiou, Athanasios Smyrli, Aikaterini Papadopoulos, Evangelos School of Mechanical Engineering National Technical University of Athens Control Systems Lab Greece

ISBN: (纸本)9798350302585

The recent years' progress in deep learning (DL) technology has resulted in convolutional neural networks (CNNs) capable of producing fast and accurate results, with minimal data preprocessing. Currently, gait analysis (GA) is attracting the attention of the field of deep learning due to its seamless integration applicability. Our approach focuses on CNN-based GA application on healthcare and orthopaedics. Using CNNs and visual fiducial systems for recognition and 3D mesh re-construction of the human form and the floor respectively, we can virtually recreate the human-floor interaction, which can be particularly useful in the study of gait dynamics, through the per-frame gait phase classification. However, the current state-of-the-art (SOTA) is that most CNN mesh reconstruction software produces a mesh from a monocular input. The use of photogrammetry could alternatively be implemented, but would require multiple cameras and expensive equipment. Our approach aims to create a refined mesh obtained from trinocular footage, along with an interactive and easy-to-use interface. © 2023 IEEE.

关键词： Floors

来源：评论

学校读者我要写书评

暂无评论

Detachment Behavior of Gecko Toe in Functional Strategies for Bionic Toe

引用

Journal of Bionic Engineering 2024年第2期21卷 707-717页

作者： Qingfei Han Wei Wang Huan Shen Xincheng Feng Haoran Zhang Qian Li Yi Sun Huapeng Wu Aihong Ji Lab of Locomotion Bioinspiration and Intelligent Robots College of Mechanical and Electrical EngineeringNanjing University of Aeronautics and AstronauticsNanjing210016China Laboratory of Intelligent Machines School of Energy SystemsLappeenranta-Lahti University of Technology53850LappeenrantaFinland State Key Laboratory of Mechanics and Control for Aerospace Structures Nanjing University of Aeronautics and AstronauticsNanjing210016China

Geckos can efficiently navigate complex terrains due to their multi-level adhesive system that is present on their *** setae are responsible for the gecko’s extraordinary adhesion and have garnered wide attention from the scientific *** majority of the reported works in the literature that have dealt with the peeling models mainly focus on the gecko hierarchical adhesive system,with limited attention given to investigating the influence of gecko toe structure on the *** these lines,to gain a deeper understanding of the rapid and effortless detachment abilities of gecko toes,the peeling behavior of gecko toes on vertical surfaces was primarily investigated in this *** specifically,the detachment time of a single toe on a smooth acrylic plate was measured to be 0.41±0.21 ***,it was observed that the toe assumed a"U"-shaped structure upon complete ***,Finite Element Analysis(FEA)models for three different types of gecko toes were developed to simulate both the displacement-peel and the moment-peel *** the segmentation of the adhesive layer led to a gradual decrease in the resultant force,as well as the normal and tangential ***,a gecko-inspired toe model was constructed and powered by Shape Memory Alloy(SMA).A systematic comparison between the vertical drag separation and the outward flip separation was also *** our analysis,it was clearly demonstrated that outward peel separation significantly necessitated the reduction of the peeling force,thus confirming the advantageous nature of the outward motion in gecko toe *** data not only contribute to a deeper understanding of the gecko detachment behavior but also offer valuable insights for the advancement of the wall-climbing robot feet.

关键词： Gecko toes Detachment behavior Finite element analysis Gecko-inspired toe

来源：评论

学校读者我要写书评

暂无评论

Preventing String Instability in Altitude control System of UAV by Nonlinear Correction 30

Preventing String Instability in Altitude Control System of ...

引用

30th Saint Petersburg International Conference on Integrated Navigation systems, ICINS 2023

作者： Zaitceva, Iuliia Andrievsky, Boris St. Petersburg State University St. Petersburg Electrotechnical University "lETI" Control of Comlex Systems Lab. Institute of Problems in Mechanical Engineering of RAS St. Petersburg Russia St. Petersburg State University Control of Comlex Systems Lab. Institute of Problems in Mechanical Engineering of RAS St. Petersburg Russia

ISBN: (纸本)9785919950950

The paper is devoted to maintaining a flight formation with cross-information links between agents while controlling the altitude of nine unmanned aerial vehicles through a standard proportional-integral-differential controller. With an increase in the number of agents, the system oscillations may appear due to the "string instability"phenomenon. Other factors associated with changes in the dynamics of the nonlinear actuator of the UAV control surface can also negatively affect the formation stability. The simulation results show that the actuator saturation for one agent can lead to oscillatory processes of other agents, which can eventually destroy the formation structure. In the paper, it is proposed to introduce a nonlinear corrective device into the actuator control loop to improve the system stability and prevent dangerous oscillations. In the nominal mode, when the saturations of actuators are not active, this corrective device does not affect the operation of the controller and passes the control signal without changes, but it makes an effect when the controlling values of actuator positions and velocities reach the bounds. In the paper, for a flight altitude control problem of the UAVs formation, the corrective device was tested and its efficiency was confirmed by a series of simulations. © 2023 Concern CSRI Elektropribor, JSC.

关键词： Altitude control

来源：评论

学校读者我要写书评

暂无评论

Bridging the Sim-to-Real Gap with Dynamic Compliance Tuning for Industrial Insertion

Bridging the Sim-to-Real Gap with Dynamic Compliance Tuning ...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Xiang Zhang Masayoshi Tomizuka Hui Li Mechanical Systems Control Lab UC Berkeley USA Autodesk Research USA

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

Contact-rich manipulation tasks often exhibit a large sim-to-real gap. For instance, industrial assembly tasks frequently involve tight insertions where the clearance is less than 0.1 mm and can even be negative when dealing with a deformable receptacle. This narrow clearance leads to complex contact dynamics that are difficult to model accurately in simulation, making it challenging to transfer simulation-learned policies to real-world robots. In this paper, we propose a novel framework for robustly learning manipulation skills for real-world tasks using simulated data only. Our framework consists of two main components: the "Force Planner" and the "Gain Tuner". The Force Planner plans both the robot motion and desired contact force, while the Gain Tuner dynamically adjusts the compliance control gains to track the desired contact force during task execution. The key insight is that by dynamically adjusting the robot’s compliance control gains during task execution, we can modulate contact force in the new environment, thereby generating trajectories similar to those trained in simulation and narrowing the sim-to-real gap. Experimental results show that our method, trained in simulation on a generic square peg-and-hole task, can generalize to a variety of real-world insertion tasks involving narrow and negative clearances, all without requiring any fine-tuning. Videos are available at https://***

关键词： Robot motion Tuners Tracking Force Dynamics Trajectory Task analysis

来源：评论

学校读者我要写书评

暂无评论

Synchronized UAV Group Flight control under Data Rate Limitations and Actuator Saturation 7

Synchronized UAV Group Flight Control under Data Rate Limita...

引用

7th Scientific School: Dynamics of Complex Networks and their Applications, DCNA 2023

作者： Andrievsky, Boris Zaitceva, Luliia St. Petersburg State University Control of Complex Systems Lab. Institute for Problems in Mechanical Engineering of Ras Department of Applied Cybernetics Saint Petersburg Russia St. Petersburg State University St. Petersburg Electrotechnical University ''LETI" Control of Complex Systems Lab. Institute for Problems in Mechanical Engineering of Ras Department of Applied Cybernetics Saint Petersburg Russia

ISBN: (纸本)9798350305746

This work specifically addresses the challenges associated with information transfer between agents in pursuit of desired control objectives. Seamless communication is crucial, especially during formation synchronization, as delays can lead to disintegration. Additionally, large-scale systems may face string or mesh instability, where state deviations increase. This work aims to advance decentralized control in multi-agent systems with complex information graph structures, considering nonlinear limitations on the control signal. A sequential nonlinear corrective unit is employed to mitigate this effect. The present study incorporates data rate limitations on information exchange between UAVs in formation. An adaptive coding/decoding procedure is utilized for data transfer, and the minimal data transmission rate required for maintaining formation stability is numerically evaluated. © 2023 IEEE.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Multicamera Visual SLAM For Vineyard Inspection

Multicamera Visual SLAM For Vineyard Inspection

引用

Mediterranean Conference on control and Automation (MED)

作者： Christos Kokas Athanasios Mastrogeorgiou Konstaninos Machairas Konstantinos Koutsoukis Evangelos Papadopoulos The Control Systems Lab (CSL) School of Mechanical Engineering National Technical University of Athens Greece

ISBN: (数字)9798350395440

ISBN: (纸本)9798350395457

Automating the process of collecting samples (e.g., images) from a vineyard can help to monitor the condition of the grapes with precision and prevent the spreading of diseases. A critical part of this task is the development of a robust localization algorithm so that (a) a robot is able to carry out the inspection process and (b) the vine-grower knows exactly which part of the vineyard has been inspected. In this paper, we propose a novel approach for enhancing the robustness of vSLAM by utilizing multiple stereo cameras and a novel method for detecting loops in homogeneous environments based on AprilTags, where state-of-the-art approaches may find it difficult to detect them. We test the accuracy of our method using a wheeled Robotic Platform (RP) in simulation and in a synthetic vineyard developed at CSL, NTUA [1]. The developed method achieves high accuracy in the localization of the RP in the vineyard and robustness even when a featureless object covers a large part of the Field of View of one camera. The developed software is available for testing at the CSL's bitbucket repository [2].

关键词： Location awareness Visualization Accuracy Robot vision systems Inspection Cameras Robustness

来源：评论

学校读者我要写书评

暂无评论

Vinymap: a Vineyard Inspection and 3D Reconstruction Framework for Agricultural Robots

Vinymap: a Vineyard Inspection and 3D Reconstruction Framewo...

引用

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

作者： Ioannis Zarras Athanasios Mastrogeorgiou Konstaninos Machairas Konstantinos Koutsoukis Evangelos Papadopoulos Control Systems Lab (CSL) School of Mechanical Engineering National Technical University of Athens Greece

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

Efficient and thorough vineyard inspection is crucial for optimizing yield and preventing disease from spreading. Manual approaches are labor-intensive and prone to human error, motivating the development of automated solutions. Precision viticulture benefits greatly from access to photo-realistic 3D vineyard maps and from capturing intricate visual details necessary for accurate canopy and grape health assessment. Generating such maps efficiently proves challenging, particularly when employing cost-effective equipment. This paper presents a novel vineyard inspection and 3D reconstruction framework implemented on a Robotic Platform (RP) equipped with three stereo cameras. The framework’s performance was evaluated on an experimental synthetic vineyard developed at NTUA. This testing setup allowed experimentation under diverse lighting conditions, ensuring the system’s robustness under realistic scenarios. Unlike existing solutions, which often focus on specific aspects of the inspection, our framework offers a top-down approach, encompassing autonomous navigation, high-fidelity 3D reconstruction, and canopy growth assessment. The developed software is available at the control systems laboratory’s (CSL) bitbucket repository [1].

关键词： Visualization Three-dimensional displays Inspection Solids Visual servoing Software Robustness Real-time systems Diseases Testing

来源：评论

学校读者我要写书评

暂无评论

Ensuring Safety for UAVs Through Event-Triggered Predictive control 32

Ensuring Safety for UAVs Through Event-Triggered Predictive ...

引用

32nd Mediterranean Conference on control and Automation, MED 2024

作者： Aspragkathos, Sotirios N. Karras, George C. Kyriakopoulos, Kostas J. School of Mechanical Engineering National Technical University of Athens Control Systems Lab Athens15780 Greece University of Thessaly Dept. of Informatics and Telecommunications Lamia35100 Greece Abu Dhabi128188 United Arab Emirates

ISBN: (纸本)9798350395440

This paper presents a control system for un-manned aerial vehicles (UAVs) that track deformable objects. Using Event-Triggered (ET) principles and Image Moments-Based Nonlinear Model Predictive control (NMPC), the suggested approach improves flexibility by triggering control changes in response to specific occurrences. The incorporation of Barrier Functions (BFs) enforces safety requirements, which are critical for negotiating uncertain terrains. The framework's implementation and testing on a UAV tracking deformable coastlines are intended to demonstrate its effectiveness in adapting to dynamic situations while adhering to safety limitations. © 2024 IEEE.

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

FDPP: Fine-tune Diffusion Policy with Human Preference

arXiv

引用

arXiv 2025年

作者： Chen, Yuxin Jha, Devesh K. Tomizuka, Masayoshi Romeres, Diego Mechanical Systems Control Lab UC Berkeley BerkeleyCA United States CambridgeMA United States

Imitation learning from human demonstrations enables robots to perform complex manipulation tasks and has recently witnessed huge success. However, these techniques often struggle to adapt behavior to new preferences or changes in the environment. To address these limitations, we propose Fine-tuning Diffusion Policy with Human Preference (FDPP). FDPP learns a reward function through preference-based learning. This reward is then used to fine-tune the pre-trained policy with reinforcement learning (RL), resulting in alignment of pre-trained policy with new human preferences while still solving the original task. Our experiments across various robotic tasks and preferences demonstrate that FDPP effectively customizes policy behavior without compromising performance. Additionally, we show that incorporating Kullback–Leibler (KL) regularization during fine-tuning prevents over-fitting and helps maintain the competencies of the initial policy. © 2025, CC BY.

关键词： Reinforcement learning

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：