检索结果-内蒙古大学图书馆

Generalizing 6-DoF Grasp Detection via Domain Prior Knowledge

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ma, Haoxiang Shi, Modi Gao, Boyang Huang, Di State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China School of Computer Science and Technology Harbin Institute of Technology Harbin China Geometry Robotics China

We focus on the generalization ability of the 6-DoF grasp detection method in this paper. While learning-based grasp detection methods can predict grasp poses for unseen objects using the grasp distribution learned from the training set, they often exhibit a significant performance drop when encountering objects with diverse shapes and structures. To enhance the grasp detection methods’ generalization ability, we incorporate domain prior knowledge of robotic grasping, enabling better adaptation to objects with significant shape and structure differences. More specifically, we employ the physical constraint regularization during the training phase to guide the model towards predicting grasps that comply with the physical rule on grasping. For the unstable grasp poses predicted on novel objects, we design a contact-score joint optimization using the projection contact map to refine these poses in cluttered scenarios. Extensive experiments conducted on the GraspNet-1billion benchmark demonstrate a substantial performance gain on the novel object set and the real-world grasping experiments also demonstrate the effectiveness of our generalizing 6-DoF grasp detection method. Code is available at https://***/mahaoxiang822/ Generalizing-Grasp. Copyright © 2024, The Authors. All rights reserved.

关键词： Benchmarking

Clustering With Adaptive Graph Learning and Spectral Rotation 7

学校读者我要写书评

暂无评论

Clustering With Adaptive Graph Learning and Spectral Rotatio...

7th International Conference on Vision, Image and Signal Processing, ICVISP 2023

作者： Yuan, Ge Zhou, Jie Huang, Chucheng Gao, Can Wang, Yangbo Shen, Xinrui National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Guangdong Shenzhen518060 China College of Computer Science and Software Engineering Shenzhen University Guangdong Shenzhen518060 China College of Mathematics and Statistics Shenzhen University Guangdong Shenzhen518060 China SZU Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society Guangdong Shenzhen518060 China

ISBN: (纸本)9781839539848

Prototype-based clustering algorithms have garnered considerable attention in the field of machine learning due to their efficiency and interpretability. Nonetheless, these algorithms often face performance degradation when confronted with high-dimensional or non-ellipsoidal data distributions. To surmount these challenges, this study introduces a novel clustering approach, dubbed Clustering with Adaptive Graph learning and Spectral Rotation (CAGSR). In CAGSR, the imposed spectral rotation operation mitigates the discrepancy between the membership matrix, which adheres to the notion of fuzzy clustering, and the spectral representations derived from an adaptive graph rather than a predefined one. This enables the generation of a comprehensive representation of the data across multiple spaces. Furthermore, the clustering and graph learning tasks are jointly optimized in a projected subspace, which can effectively reduce the adverse impact caused by irrelevant features in the original space. The proposed method seamlessly integrates fuzzy clustering, graph structure learning, and spectral rotation into a unified model, facilitating the detection of intrinsic structures. Experimental evaluations conducted on benchmark data sets substantiate the effectiveness of CAGSR when compared to related clustering approaches. © The Institution of engineering & Technology 2023.

关键词： Fuzzy clustering

Motion planning for highly-dynamic unconditioned reflexes based on chained Signed Distance Functions

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Lin, Ken Ye, Qi Lam, Tin Lun Li, Zhibin Chen, Jiming Li, Gaofeng College of Control Science and Engineering Zhejiang University Hangzhou China Robotics and Artificial Intelligence Laboratory The Chinese University of Hongkong Shenzhen China Department of Computer Science University College London London United Kingdom

The unconditioned reflex (e.g., protective reflex), which is the innate reaction of the organism and usually performed through the spinal cord rather than the brain, can enable organisms to escape harms from environments. Accordingly, it is essential to endow robots the highly-dynamic unconditioned reflexes to humans and/or environments, such that it can work collaboratively with humans or serve people in daily life. Although various sampling-based and optimization-based motion planning methods have been developed for robots, their limited computational speed is not suitable for rapid motion planning in highly-dynamic environments. In this paper, we propose an online, highly-dynamic motion planning algorithm to enable highly-dynamic unconditioned reflexes for manipulators. Our method is based on a chained version of Signed Distance Functions (SDFs), which can be pre-computed and stored. Our proposed algorithm is divided into two stages. In the offline stage, we create 3 groups of local SDFs to store the geometric information of the manipulator and its working environment. The SDFs in the first group are to describe the geometric information of the static environment. The SDFs in the second group are to describe the geometric information of each link. While the SDFs in the last group are to describe the reachability of each link’s end of the manipulator. In the online stage, the precomputed local SDFs are chained together according the configuration of the manipulator, to provide global geometric information about the environment. While the point clouds of the dynamic objects serve as query points to look up these local SDFs for quickly generating escape velocity. Then we propose a modified geometric Jacobian matrix and use the Jacobian-pseudo-inverse method to generate real-time reflex behaviors to avoid the static and dynamic obstacles in the environment. The benefits of our method are validated in both static and dynamic scenarios. In the static scenario, our meth

关键词： Jacobian matrices

GM-DF: Generalized Multi-Scenario Deepfake Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lai, Yingxin Yu, Zitong Yang, Jing Li, Bin Kang, Xiangui Shen, Linlin The School of Computing and Information Technology Great Bay University Dongguan523000 China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen University Shenzhen518060 China The Guangdong Key Laboratory of Information Security The School of Computer Science and Engineering Sun Yat-sen University Guangzhou510080 China Computer Vision Institute School of Computer Science & Software Engineering Shenzhen Institute of Artificial Intelligence and Robotics for Society Guangdong Key Laboratory of Intelligent Information Processing National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China

Existing face forgery detection usually follows the paradigm of training models in a single domain, which leads to limited generalization capacity when unseen scenarios and unknown attacks occur. In this paper, we elaborately investigate the generalization capacity of deepfake detection models when jointly trained on multiple face forgery detection datasets. We first find a rapid degradation of detection accuracy when models are directly trained on combined datasets due to the discrepancy across collection scenarios and generation methods. To address the above issue, a Generalized Multi-Scenario Deepfake Detection framework (GM-DF) is proposed to serve multiple real-world scenarios by a unified model. First, we propose a hybrid expert modeling approach for domain-specific real/forgery feature extraction. Besides, as for the commonality representation, we use CLIP to extract the common features for better aligning visual and textual features across domains. Meanwhile, we introduce a masked image reconstruction mechanism to force models to capture rich forged details. Finally, we supervise the models via a domain-aware meta-learning strategy to further enhance their generalization capacities. Specifically, we design a novel domain alignment loss to strongly align the distributions of the meta-test domains and meta-train domains. Thus, the updated models are able to represent both specific and common real/forgery features across multiple datasets. In consideration of the lack of study of multi-dataset training, we establish a new benchmark leveraging multi-source data to fairly evaluate the models' generalization capacity on unseen scenarios. Both qualitative and quantitative experiments on five datasets conducted on traditional protocols as well as the proposed benchmark demonstrate the effectiveness of our approach. The codes will be available on https://***/laiyingxin2/GM-DF. Copyright © 2024, The Authors. All rights reserved.

关键词： Image reconstruction

OneTracker: Unifying Visual Object Tracking with Foundation Models and Efficient Tuning

学校读者我要写书评

暂无评论

OneTracker: Unifying Visual Object Tracking with Foundation ...

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Lingyi Hong Shilin Yan Renrui Zhang Wanyun Li Xinyu Zhou Pinxue Guo Kaixun Jiang Yiting Chen Jinglun Li Zhaoyu Chen Wenqiang Zhang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China CUHK MMLab Shanghai Artificial Intelligence Laboratory Shanghai Engineering Research Center of AI & Robotics Academy for Engineering & Technology Fudan University Shanghai China Engineering Research Center of AI & Robotics Ministry of Education Academy for Engineering & Technology Fudan University Shanghai China

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Visual object tracking aims to localize the target object of each frame based on its initial appearance in the first frame. Depending on the input modility, tracking tasks can be divided into RGB tracking and RGB+X (e.g. RGB+N, and RGB+D) tracking. Despite the different input modalities, the core aspect of tracking is the temporal matching. Based on this common ground, we present a general framework to unify various tracking tasks, termed as One Tracker. One- Tracker first performs a large-scale pre-training on a RGB tracker called Foundation Tracker. This pretraining phase equips the Foundation Tracker with a stable ability to estimate the location of the target object. Then we regard other modality information as prompt and build Prompt Tracker upon Foundation Tracker. Through freezing the Foundation Tracker and only adjusting some additional trainable parameters, Prompt Tracker inhibits the strong localization ability from Foundation Tracker and achieves parameter- efficient finetuning on downstream RGB+X tracking tasks. To evaluate the effectiveness of our general framework OneTracker, which is consisted of Foundation Tracker and Prompt Tracker, we conduct extensive experiments on 6 popular tracking tasks across 11 benchmarks and our One- Tracker outperforms other models and achieves state-of-the-art performance.

关键词： Location awareness Visualization computer vision Adaptation models Target tracking Benchmark testing Pattern recognition

Bimanual Manipulation of Steady Hand Eye Robots with Adaptive Sclera Force Control: Cooperative vs. Teleoperation Strategies

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Esfandiari, Mojtaba Gehlbach, Peter Taylor, Russell H. Iordachita, Iulian I. The Department of Mechanical Engineering The Laboratory for Computational Sensing and Robotics The Johns Hopkins University BaltimoreMD21218 United States The Wilmer Eye Institute Johns Hopkins Hospital BaltimoreMD21287 United States The Department of Computer Science The Laboratory for Computational Sensing and Robotics The Johns Hopkins University BaltimoreMD21218 United States

Performing retinal vein cannulation (RVC) as a potential treatment for retinal vein occlusion (RVO) without the assistance of a surgical robotic system is very challenging to do safely. The main limitation is the physiological hand tremor of surgeons. Robot-assisted eye surgery technology may resolve the problems of hand tremors and fatigue and improve the safety and precision of RVC. The Steady-Hand Eye Robot (SHER) is an admittance-based robotic system that can filter out hand tremors and enables ophthalmologists to manipulate a surgical instrument inside the eye cooperatively. However, the admittance-based cooperative control mode does not safely minimize the contact force between the surgical instrument and the sclera to prevent tissue damage. Additionally, features like haptic feedback or hand motion scaling, which can improve the safety and precision of surgery, require a teleoperation control framework. This work presents a bimanual adaptive teleoperation (BMAT) control framework using SHER 2.0 and SHER 2.1 robotic systems. We integrate them with an adaptive force control (AFC) algorithm to automatically minimize the tool-sclera interaction force. The scleral forces are measured using two fiber Bragg grating (FBG)-based force-sensing tools. We compare the performance of the BMAT mode with a bimanual adaptive cooperative (BMAC) mode in a vessel-following experiment under a surgical microscope. Experimental results demonstrate the effectiveness of the proposed BMAT control framework in performing a safe bimanual telemanipulation of the eye without over-stretching it, even in the absence of registration between the two robots. © 2024, CC BY.

关键词： Fiber Bragg gratings

ODD: Omni Differential Drive for Simultaneous Reconfiguration and Omnidirectional Mobility of Wheeled Robots

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhao, Ziqi Xie, Peijia Meng, Max Q.-H. Shenzhen Key Laboratory of Robotics Perception and Intelligence The Department of Electronic and Electrical Engineering Southern University of Science and Technology Shenzhen China The Department of Electronic Engineering The Chinese University of Hong Kong Hong Kong The Department of Electrical and Computer Engineering The University of Alberta Canada

Wheeled robots are highly efficient in human living environments. However, conventional wheeled designs, with their limited degrees of freedom and constraints in robot configuration, struggle to simultaneously achieve stability, passability, and agility due to varying footprint needs. This paper proposes a novel robot drive model inspired by human movements, termed as the Omni Differential Drive (ODD). The ODD model innovatively utilizes a lateral differential drive to adjust wheel spacing without adding additional actuators to the existing omnidirectional drive. This approach enables wheeled robots to achieve both simultaneous reconfiguration and omnidirectional mobility. To validate the feasibility of the ODD model, a functional prototype was developed, followed by comprehensive kinematic analyses. Control systems for self-balancing and motion control were designed and implemented. Experimental validations confirmed the feasibility of the ODD mechanism and the effectiveness of the control strategies. The results underline the potential of this innovative drive system to enhance the mobility and adaptability of robotic platforms. Copyright © 2024, The Authors. All rights reserved.

关键词： Kinematics

Learning a Better Control Barrier Function

学校读者我要写书评

暂无评论

Learning a Better Control Barrier Function

IEEE Conference on Decision and Control

作者： Bolun Dai Prashanth Krishnamurthy Farshad Khorrami Electrical & Computer Engineering Department Control/Robotics Research Laboratory Tandon School of Engineering New York University Brooklyn NY

ISBN: (数字)9781665467612

ISBN: (纸本)9781665467629

Control barrier functions (CBFs) are widely used in safety-critical controllers. However, constructing a valid CBF is challenging, especially under nonlinear or non-convex constraints and for high relative degree systems. Meanwhile, finding a conservative CBF that only recovers a portion of the true safe set is usually possible. In this work, starting from a "conservative" handcrafted CBF (HCBF), we develop a method to find a CBF that recovers a reasonably larger portion of the safe set. Since the learned CBF controller is not guaranteed to be safe during training iterations, we use a model predictive controller (MPC) to ensure safety during training. Using the collected trajectory data containing safe and unsafe interactions, we train a neural network to estimate the difference between the HCBF and a CBF that recovers a closer solution to the true safe set. With our proposed approach, we can generate safe controllers that are less conservative and computationally more efficient. We validate our approach on two systems: a second-order integrator and a ball-on-beam.

关键词： Training Heuristic algorithms Neural networks Predictive models Data collection Data models Safety

Hybrid polynomial-based trajectory planning for lower limb exoskeleton robots 18

学校读者我要写书评

暂无评论

Hybrid polynomial-based trajectory planning for lower limb e...

18th IEEE Conference on Industrial Electronics and Applications, ICIEA 2023

作者： Li, Shupei Zhang, Chi Yi, Feng Lv, Pingping Yuan, Ting Zhang, Meitong University of Chinese Academy of Sciences Ningbo Institute of Materials Technology and Engineering Chinese Academy of Sciences College of Materials Science and Opto-Electronic Technology Ningbo315201 China Ningbo Institute of Materials Technology and Engineering Zhejiang Key Laboratory of Robotics and Intelligent Manufacturing Equipment Technology Ningbo315201 China Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo315201 China Hebei University of Science & Technology School of Electrical Engineering Ningbo315201 China

ISBN: (纸本)9798350312201

To ensure smooth, continuous, and precise motion of lower limb exoskeleton robots, this paper proposes a 6-5-6 polynomial trajectory planning method (6-5-6 PTPM) that combines 5th and 6th polynomials. By imposing corresponding kinematic constraints, this method achieves the trajectory plan of seven sub-phases of the gait cycle on the kinematic model of the lower limb exoskeleton robot. The simulation is conducted and compared to the 4-3-4 polynomial trajectory planning method (4-3-4 PTPM). The comparison results indicate that the 6-5-6 PTPM is shown to have a lower error between the planned trajectory and the actual trajectory. Additionally, it avoids spikes of the acceleration curve that existed in the 4-3-4 polynomial method. The proposed method enables the acceleration curve smoother and mechanical impact lessened, which demonstrates the feasibility and stability of the proposed method. © 2023 IEEE.

关键词： Trajectories