检索结果-内蒙古大学图书馆

Dexterous Pre-grasp Manipulation for Human-like Functional Categorical Grasping: Deep Reinforcement Learning and Grasp Representations

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Pavlichenko, Dmytro Behnke, Sven Group Computer Science Institute VI – Intelligent Systems and Robotics The Center for Robotics The Lamarr Institute for Machine Learning and Artificial Intelligence University of Bonn Germany

Many objects, such as tools and household items, can be used only if grasped in a very specific way—grasped functionally. Often, a direct functional grasp is not possible, though. We propose a method for learning a dexterous pre-grasp manipulation policy to achieve human-like functional grasps using deep reinforcement learning. We introduce a dense multi-component reward function that enables learning a single policy, capable of dexterous pre-grasp manipulation of novel instances of several known object categories with an anthropomorphic hand. The policy is learned purely by means of reinforcement learning from scratch, without any expert demonstrations. It implicitly learns to reposition and reorient objects of complex shapes to achieve given functional grasps. In addition, we explore two different ways to represent a desired grasp: explicit and more abstract, constraint-based. We show that our method consistently learns to successfully manipulate and achieve desired grasps on previously unseen object instances of known categories using both grasp representations. Training is completed on a single GPU in under three hours. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep reinforcement learning

FT-TF: A 4D Long-term Flight Trajectory Prediction Method Based on Transformer

学校读者我要写书评

暂无评论

FT-TF: A 4D Long-term Flight Trajectory Prediction Method Ba...

第42届中国控制会议

作者： Shijin Luo Minghui Zhao Zhenjie Zhao Lun Li Shiyong Zhang Xuebo Zhang Institute of Robotics and Automatic Information System College of Artificial Intelligence Nankai University Tianjin Key Laboratory of Intelligent Robotics Nankai University

The accurate prediction of target aircraft trajectory in the process of air combat can significantly improve the ability of aircraft to gain air superiority. Most of the trajectory prediction methods currently applied in air combat are based on traditional time-series prediction algorithms such as LSTM with short prediction steps, which can not realize the demand for long-term prediction in air combat. To address the issue, a novel prediction method FT-TF is proposed in this paper. FT-TF is a multi-step4D flight trajectory prediction method based on Transformer applied to further improve the accuracy and practicality of trajectory prediction problems. FT-TF uses the first 30 steps of aircraft historical trajectory data as input and the last 30 steps as *** trajectory data sets of aircraft in air combat are used in the experiment to make the trained model closer to the real air combat scene. The experimental results show that the proposed FT-TF can achieve higher prediction accuracy than LSTM and BP neural network. Related Transformer variants of the algorithm are also added to the experiment for comparison, the results demonstrate that the proposed method achieves the smallest prediction error.

关键词：

Optimized Disturbance Rejection Control for Flutter Suppression of a Benchmark Active Control Technology Wind-Tunnel Model

学校读者我要写书评

暂无评论

Optimized Disturbance Rejection Control for Flutter Suppres...

International Conference on Guidance, Navigation and Control, ICGNC 2022

作者： Piao, Minnan Li, Haifeng Lu, Biao Wang, Yongshuai Huang, Jian College of Computer Science and Technology Civil Aviation University of China Tianjin300300 China Institute of Robotics and Automatic Information System College of Artificial Intelligence Nankai University Tianjin300350 China Beijing Automatic Control & Equipment Institute Beijing100074 China

ISBN: (纸本)9789811966125

Active control of aeroservoelasticity effect is critical for future aircraft. This paper concerns the flutter suppression problem for a benchmark active control technology (BACT) wind-tunnel model, which is subject to unmodelled dynamics, parameter perturbations, turbulence disturbances, and measurement noises. After analyzing the characteristics of the BACT model, the trailing edge velocity, which combines the information of the plunge and the pitch modes, is chosen as the controlled output. Several practical filters are employed to effectively attenuate the sensor noise and drift. Based on the filtered signal, we employ the active disturbance rejection control (ADRC) to compensate the multifarious uncertainties and achieve the flutter suppression. Furthermore, to facilitate parameter tuning, the nonsmooth H∞ synthesis method is employed. Finally, extensive simulation and comparison results validate the superiority of the proposed method. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Disturbance rejection

ALCHEMY: AMPLIFYING THEOREM-PROVING CAPABILITY THROUGH SYMBOLIC MUTATION

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wu, Shaonan Lu, Shuai Gong, Yeyun Duan, Nan Wei, Ping National Key Laboratory of Human-Machine Hybrid Augmented Intelligence China Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University China Microsoft Research Asia China

Formal proofs are challenging to write even for experienced experts. Recent progress in Neural Theorem Proving (NTP) shows promise in expediting this process. However, the formal corpora available on the Internet are limited compared to the general text, posing a significant data scarcity challenge for NTP. To address this issue, this work proposes Alchemy, a general framework for data synthesis that constructs formal theorems through symbolic mutation. Specifically, for each candidate theorem in Mathlib, we identify all invocable theorems that can be used to rewrite or apply to it. Subsequently, we mutate the candidate theorem by replacing the corresponding term in the statement with its equivalent form or antecedent. As a result, our method increases the number of theorems in Mathlib by an order of magnitude, from 110k to 6M. Furthermore, we perform continual pretraining and supervised finetuning on this augmented corpus for large language models. Experimental results demonstrate the effectiveness of our approach, achieving a 5% absolute performance improvement on Leandojo benchmark. Additionally, our synthetic data achieve a 2.5% absolute performance gain on the out-of-distribution miniF2F benchmark. To provide further insights, we conduct a comprehensive analysis of synthetic data composition and the training paradigm, offering valuable guidance for developing a strong theorem prover. Copyright © 2024, The Authors. All rights reserved.

关键词：

Recruiting Heterogeneous Crowdsource Vehicles for Updating a High-Definition Map

学校读者我要写书评

暂无评论

Recruiting Heterogeneous Crowdsource Vehicles for Updating a...

International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt)

作者： Wentao Ye Yuan Luo Bo Liu Jianwei Huang Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS) School of Science and Engineering The Chinese University of Hong Kong Shenzhen Guangdong China Guangdong Provincial Key Laboratory of Aerospace Communication and Networking Technology School of Science and Engineering Shenzhen Institute of Artificial Intelligence and Robotics for Society The Chinese University of Hong Kong Shenzhen Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS) Guangdong China School of Science and Engineering Shenzhen Institute of Artificial Intelligence and Robotics for Society The Chinese University of Hong Kong Shenzhen Shenzhen China

The high-definition map is a cornerstone of autonomous driving. Unlike constructing a costly fleet of mapping vehicles, the crowdsourcing paradigm is a cost-effective way to keep an HD map up to date. Achieving practical success for crowdsourcing-based HD maps is contingent on addressing two critical issues: freshness and recruitment costs. Given that crowdsource vehicles are often heterogeneous in terms of operational costs and sensing capabilities, it is practical to recruit heterogeneous crowdsource vehicles to achieve the tradeoff between freshness and recruitment costs. However, existing works neglect this aspect. To solve it, we formulate this problem as a Markov decision process. We demonstrate that the optimal policy is threshold-type age-dependent. Additionally, our findings reveal some counter-intuitive insights. In some cases, the company should initiate vehicle recruitment earlier when vehicles arrive more frequently, or have higher operational costs or sensing capabilities. Besides, we propose an efficient algorithm, called the bound-based relative value iteration (BRVI) algorithm, to overcome the technical challenge that finding an optimal policy is time-consuming. Numerical simulations show that (i) the optimal policy reduces the average cost by 19.04℅ compared to the state-of-the-art mechanism, and (ii) the proposed algorithm can reduce the convergence time by 13.66℅ on average compared to the existing algorithm.

关键词：

Evaluation of Continual Learning Methods for Object Hardness Recognition

学校读者我要写书评

暂无评论

Evaluation of Continual Learning Methods for Object Hardness...

2023 IEEE International Conference on Real-Time Computing and robotics, RCAR 2023

作者： Liu, Yiwen Fang, Senlin Liu, Chengliang Wang, Jingnan Mai, Ke Zhang, Yupo Yi, Zhengkun Wu, Xinyu Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Provincial Key Laboratory of Robotics and Intelligent System Shenzhen518055 China University of Chinese Academy of Sciences Beijing100049 China Shenzhen Institute of Artificial Intelligence and Robotics for Society Siat Branch Shenzhen518055 China City University of Macau Faculty of Data Science 999078 China

ISBN: (纸本)9798350327182

How to solve the problem of continual learning of robotic tactile perception in an open, dynamic environment is a pressing task. Although continual learning has been extensively studied in computer vision, the evaluation of those algorithms does not provide insights on whether their solutions may help continual learning in the context of tactile robotics. In this paper, we propose a continual learning scenario in the robotic tactile domain, and design experiments with tactile hardness dataset that allow us to compare the performance of many continual learning algorithms in the tactile domain. We examine the performance of regularization-based and replay methods in the tactile domain and discuss the effect of replay number on model performance. We show that regularization-based methods do not perform well on the tactile dataset. On the contrary, replay methods can maintain better anti-forgetting performance, but the number of model parameters will be larger, and the training time will be longer. We also found that the replay ratio should be higher than 0.2 to maintain the performance of the replay method on the tactile hardness dataset. © 2023 IEEE.

关键词： robotics

Scalable and eco-friendly flexible loudspeakers for distributed human-machine interactions

学校读者我要写书评

暂无评论

npj Flexible Electronics 2023年第1期7卷 89-98页

作者： Yucong Pi Qiutong Liu Zhaoyang Li Dazhe Zhao Kaijun Zhang Zhirui Liu Bingpu Zhou Iek Man Lei Yuan Ma Junwen Zhong Department of Electromechanical Engineering and Centre for Artificial Intelligence and Robotics University of MacaoMacao SAR 999078China Department of Mechanical Engineering and Research Institute for Intelligent Wearable Systems The Hong Kong Polytechnic UniversityHong KongChina Joint Key Laboratory of the Ministry of Education Institute of Applied Physics and Materials EngineeringUniversity of MacaoAvenida da UniversidadeTaipaMacao 999078China

Flexible loudspeakers that can be easily distributed in the surrounding environment are essential for creating immersive experiences in human-machine interactions,as these devices can transmit acoustic information *** this paper,we present a flexible electret loudspeaker that offers numerous benefits,such as eco-friendly,easy fabrication,flexible customization,strong durability,and excellent *** output sound pressure level(SPL)and frequency response characteristic are optimized according to the simulation and experiment *** a distance of 50 meters,a large-size loudspeaker(50×40 cm^(2))can produce an average SPL of 60 dB(normal SPL range of human voices is between 40 to 70 dB).The frequency response of our loudspeaker is high and relatively consistent up to 15 kHz,which covers the normal frequency range of human voices(<8 kHz).As demonstrated in this work,our loudspeakers can be used for scalable applications,such as being integrated with curtains or hung up like posters,offering a promising and practical solution for creating better human-machine interaction experiences.

关键词： speakers durability offering

UAV Obstacle Avoidance by Human-in-the-Loop Reinforcement in Arbitrary 3D Environment

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Li, Xuyang Fang, Jianwu Du, Kai Mei, Kuizhi Xue, Jianru The Chang'an University Xi'An China The Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'An China

This paper focuses on the continuous control of the unmanned aerial vehicle (UAV) based on a deep reinforcement learning method for a large-scale 3D complex environment. The purpose is to make the UAV reach any target point from a certain starting point, and the flying height and speed are variable during navigation. In this work, we propose a deep reinforcement learning (DRL)-based method combined with human-in-the-loop, which allows the UAV to avoid obstacles automatically during flying. We design multiple reward functions based on the relevant domain knowledge to guide UAV navigation. The role of human-in-the-loop is to dynamically change the reward function of the UAV in different situations to suit the obstacle avoidance of the UAV better. We verify the success rate and average step size on urban, rural, and forest scenarios, and the experimental results show that the proposed method can reduce the training convergence time and improve the efficiency and accuracy of navigation tasks. The code is available on the website https://***/Monnalo/UAV_navigation. Copyright © 2023, The Authors. All rights reserved.

关键词： Reinforcement learning

Technologies on Effectiveness and Efficiency: A Survey of State Spaces Models

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Lv, Xingtai Sun, Youbang Zhang, Kaiyan Qu, Shang Zhu, Xuekai Fan, Yuchen Wu, Yi Hua, Ermo Long, Xinwei Ding, Ning Zhou, Bowen Department of Electronic Engineering Tsinghua University China Shanghai Artificial Intelligence Laboratory China Robotics Institute Carnegie Mellon University United States

State Space Models (SSMs) have emerged as a promising alternative to the popular transformer-based models and have been increasingly gaining attention. Compared to transformers, SSMs excel at tasks with sequential data or longer contexts, demonstrating comparable performances with significant efficiency gains. In this survey, we provide a coherent and systematic overview for SSMs, including their theoretical motivations, mathematical formulations, comparison with existing model classes, and various applications. We divide the SSM series into three main sections, providing a detailed introduction to the original SSM, the structured SSM represented by S4, and the selective SSM typified by Mamba. We put an emphasis on technicality, and highlight the various key techniques introduced to address the effectiveness and efficiency of SSMs. We hope this manuscript serves as an introduction for researchers to explore the theoretical foundations of SSMs. Copyright © 2025, The Authors. All rights reserved.

关键词： State space methods