检索结果-内蒙古大学图书馆

Reinforcement Learning for Highway Driving Decision-making Using CDR (Curriculum based Domain Randomization)

Journal of Institute of control, robotics and Systems 2025年第2期31卷 114-123页

作者： Yong, Jae-Hyoung Kim, Min-Sung Park, Tae-Hyoung Department of Intelligent System and Robotics Chungbuk National University Korea Republic of Department of Control Robot Engineering Chungbuk National University Korea Republic of Department of Intelligent System and Robotics Chungbuk National University Chungbuk National University Hospital Korea Republic of

Incorrect autonomous driving decisions on highways can lead to traffic congestion and accidents. Therefore, accurate decision-making in highways is essential. However, decision-making in highways is a challenging task due to the complexity of traffic situations. Thus, reinforcement learning is an appropriate approach for decision-making in highways. However, a domain gap often arises between the training and testing environments due to differences in them. Therefore, a curriculum-based domain randomization approach is applied, which progressively adjusts the level of random parameters. The RL model is trained using the SUMO simulator and tested in the MORAI simulator, where white noise is added to simulate real-world conditions. Consequently, applying curriculum domain randomization has the lowest collision rate. © ICROS 2025.

关键词： Traffic congestion

来源：评论

学校读者我要写书评

暂无评论

Camera-LiDAR Fusion for Detecting Traffic Law Violations in Surveillance Systems

引用

Journal of Institute of control, robotics and Systems 2025年第3期31卷 187-195页

作者： Kim, Taek-Lim Jang, Byoung-Jin Yeon, Joo-Yeon Park, Tae Hyoung Department of Control and Robot Engineering Chungbuk National University Korea Republic of Department of Intelligent System and Robotics Chungbuk National University Korea Republic of

Traffic surveillance systems are essential for ensuring public safety and optimizing urban traffic flow by accurately detecting, classifying, and monitoring traffic law violations such as illegal parking and jaywalking. However, detecting such violations in dynamic urban environments poses significant challenges due to unpredictable conditions. Traditional camera-based systems excel in capturing visual details but suffer performance degradation under low light, occlusion, or complex intersection scenarios, leading to false positives and missed detections. Therefore, we propose a camera-LiDAR fusion-based traffic violation detection system. LiDAR's independence from lighting conditions complements camera capabilities, especially in night scenarios or occluded environments. Experimental results in a real-world urban environment demonstrate the system’s effectiveness in reducing false positives and missed detections compared to traditional camera-only approaches. This study underscores the potential of sensor fusion to improve traffic management systems and points out future research opportunities in challenging conditions. © ICROS 2025.

关键词： Urban transportation

来源：评论

学校读者我要写书评

暂无评论

Image Segmentation System of Soft Capsules Based on Deep Learning

引用

Journal of Institute of control, robotics and Systems 2024年第6期30卷 607-613页

作者： Jung, Young-Gil Park, Tae-Hyoung Department of Control and Robot Engineering Chungbuk National University Korea Republic of Department of Intelligent Systems and Robotics Chungbuk National University Korea Republic of

Pharmacists in conventional pharmacies and hospitals prepare and visually inspect medications before dispensing these medications to patients, thereby posing risks of exposure through skin contact or inhalation, as well as the potential for generating compounding errors such as incorrect medication types or quantities. To address the limitations of the manual compounding processes, automated compounding and medication-verification systems have been introduced and implemented. The automated compounding system prepares medications in pouch units based on prescriptions written by doctors, whereas the medication verification system verifies whether the types and quantities of the medication dispensed in these pouches match the prescription information. Medications are categorized broadly into solid tablets, hard capsules, and soft capsules. Color and shape information is used to classify medication types;solid tablets and hard capsules can be classified using rule-based image processing techniques such as image binarization. However, extracting shape information from soft capsules through which light passes is impossible via image binarization alone;therefore pattern matching is employed. Nevertheless, soft capsules change their forms, causing significant variations in pattern-matching similarity;therefore the detection accuracies of these capsules are low. To address this issue, this paper proposes a deep learning-based method for detecting soft capsules. © ICROS 2024.

关键词： Pattern matching

来源：评论

学校读者我要写书评

暂无评论

Learning Agile Quadrotor Flight in Restricted Environments with Safety Guarantees

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on intelligent Vehicles 2024年第9期9卷 1-8页

作者： Chen, Shiyu Li, Yanjie Lou, Yunjiang Lin, Ke Wu, Xinyu Department of Control Science and Engineering Harbin Institute of Technology Shenzhen and Guangdong Key Laboratory of Intelligent Morphing Mechanisms and Adaptive Robotics Shenzhen China Guangdong Provincial Key Laboratory of Robotics and Intelligent System Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China

With the increasing requirement for agile and efficient controllers in safety-critical scenarios, controllers that exhibit both agility and safety are attracting attention, especially in the aerial robotics domain. This paper focuses on the safety issue of Reinforcement Learning (RL)-based control for agile quadrotor flight in restricted environments. To this end, we propose a unified Adaptive Safety Predictive Corrector (ASPC) to certify each output action of the RL-based controller in real-time, ensuring its safety while maintaining agility. Specifically, we develop the ASPC as a finite-horizon optimal control problem, formulated by a variant of Model Predictive control (MPC). Given the safety constraints determined by the restricted environment, the objective of minimizing loss of agility can be optimized by reducing the difference between the actions of RL and ASPC. As the safety constraints are decoupled from the RL-based control policy, the ASPC is plug-and-play and can be incorporated into any potentially unsafe controllers. Furthermore, an online adaptive regulator is presented to adjust the safety bounds of the state constraints with respect to the environment changes, extending the proposed ASPC to different restricted environments. Finally, simulations and real-world experiments are demonstrated in various restricted environments to validate the effectiveness of the proposed ASPC. IEEE

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Identification of New EMC Factors of UAV FHSS Datalink in the Pre-Compliance Test Phase

引用

Journal of Engineering Science and Technology Review 2023年第3期16卷 109-116页

作者： Imrane, Bennageh Hassane, Mahmoudi Department of power electronics intelligent control techniques automation and robotics Mohammadia School of Engineers Mohammed V University of Rabat Morocco

This document presents a cost-effective methodology to diagnose the various parameters that influence the electromagnetic susceptibility (EMS) of a datalink using the frequency hopping transmission technique in drone applications. The innovative method is applied during the electromagnetic compatibility (EMC) pre-compliance test of a drone datalink. The paper outlines essential guidelines for establishing a reliable and affordable EMC test bench setup. It focuses on modifying specific parameters of the frequency hopping datalink that are typically not considered in conventional EMC testing. The results obtained reveal important new parameters of the datalink that utilizes hopping spread spectrum transmission techniques, including the communication protocol, datalink frequency bandwidth, and the number of transmission channels. By manipulating these key parameters, the electromagnetic immunity of the datalink used in experiments significantly improves. © 2023 School of Science, IHU. All rights reserved.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Aggressive and robust low-level control and trajectory tracking for quadrotors with deep reinforcement learning

引用

Neural Computing and Applications 2025年第3期37卷 1223-1240页

作者： Chen, Shiyu Li, Yanjie Lou, Yunjiang Lin, Ke Department of Control Science and Engineering Harbin Institute of Technology Shenzhen Shenzhen518055 China Guangdong Key Laboratory of Intelligent Morphing Mechanisms and Adaptive Robotics Shenzhen518055 China

Executing accurate trajectory tracking tasks using a high-performance low-level controller is crucial for quadrotors to be applied in various scenarios, especially those involving uncertain disturbances. However, due to the uncertainties in disturbed environments, developing effective low-level controllers with traditional model-based control schemes is challenging. This paper presents an aggressive and robust reinforcement learning (RL)-based low-level control policy for quadrotors. The policy maps the observed quadrotor state directly to motor thrust commands, without requiring the quadrotor dynamics. Additionally, a trajectory generation pipeline is developed to improve the accuracy of trajectory tracking tasks based on differential flatness. With the learned low-level control policy, extensive simulations and real-world experiments are implemented to validate the performance of the policy. The results indicate that our RL-based low-level control policy outperforms traditional proportional–integral–derivative (PID) control methods and related learning-based policies in terms of accuracy and robustness, particularly in environments with uncertain disturbances. Furthermore, the proposed RL-based control policy exhibits an aggressive response in trajectory tracking, even when the speed of the desired trajectory is increased to 6 m/s. Moreover, the learned policy demonstrates strong vibration suppression capabilities and enables the quadrotor to recover to a hovering state from random initial conditions with shorter response time. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

All-Fabric Bi-directional Actuators for Multi-joint Assistance of Upper Limb

引用

Journal of Bionic Engineering 2023年第6期20卷 2661-2669页

作者： Junlin Ma Diansheng Chen Zhe Liu Jie Wei Xianglin Zhang Zihan Zeng Yongkang Jiang Institute of Robotics Beihang UniversityBeijing100191China Department of Control Science and Engineering Tongji UniversityShanghai201804China Frontiers Science Center for Intelligent Autonomous Systems Shanghai201210China

According to clinical studies,upper limb robotic suits are vital to reduce therapist fatigue and accelerate patient *** pneumatic actuators have drawn increasing attention for the development of wearable robots due to their low weight,flexibility,and high power-to-weight ***,most of current actuators were designed for the flexion assistance of a specific joint,and that for joint extension requires further ***,designing an actuator for diverse working scenarios remains a *** this paper,we propose an all-fabric bi-directional actuator to assist the flexion and extension of the elbow,wrist,and fingers.A mathematical model is presented that predicts the deformation and guides the design of the proposed bi-directional *** further validate the applicability and adaptability of the proposed actuator for different joints,we developed a 3-DOF soft robotic *** results show that the robotic suit can assist the motion of the elbow,wrist,and finger of the subject.

关键词： Soft robotic suits Fabric-based actuators Multi-DOF motion Upper limb rehabilitation

来源：评论

学校读者我要写书评

暂无评论

Transfer Learning: An Approach on the Riemannian Manifold for Multiclass Motor Imagery EEG Brain-Computer Interface 30

Transfer Learning: An Approach on the Riemannian Manifold fo...

引用

30th International Conference on Mechatronics and Machine Vision in Practice, M2VIP 2024

作者： Sun, Xu Chu, Yaqi Luo, Yang Zhao, Xingang Northeastern University Department of Control Engineering Shenyang110819 China China Institutes for Robotics and Intelligent Manufacturing Shenyang110016 China

ISBN: (纸本)9798350391916

Motor imagery (MI) EEG-based brain-computer interface (BCI) facilitates direct communication between the brain's intentions and a computer. In essence, it allows a person to control external equipment by decoding EEG signals during the mental imagery of limb movements. However, due to the intricate individual variances in EEG signals, creating a general decoding model with parameters applicable to all subjects is exceptionally challenging. Consequently, a prolonged calibration process is necessary to gather labeled subject-specific data for each individual, making it less user-friendly. A potential solution to this problem is transfer learning, a methodology that transfers knowledge from related domains to a target domain. To tackle the problem, this paper proposes a novel transfer learning approach on the Riemannian manifold framework in the context of multiclass MI EEG-based BCI classification. Particularly, a user-specific frequency band selection (FBS) method with MI EEG class distinctiveness is utilized to improve the accuracy and efficiency of Riemannian space calculations, which is measured using the inter-class distance and intra-class variance on the manifold. Then, the Riemannian space Alignment (RA) strategy is used to calibrate the MI EEG variances of different subjects. The comparative experiments between the proposed approach and baseline/conventional methods are conducted on a public dataset with four-class motor imagery EEG, including two-class transfer learning scenario and four-class transfer learning scenario. proposed transfer learning approach is outperformed. Compared with the baseline/conventional methods using a fixed wide frequency band, the preliminary results suggests that the proposed approach can significantly improve the transfer learning performance for the MI EEG signals from different subjects in Riemannian space. © 2024 IEEE.

关键词： Brain computer interface

来源：评论

学校读者我要写书评

暂无评论

Long-Range Vehicle Detection of LiDAR using Adaboost Classification

Long-Range Vehicle Detection of LiDAR using Adaboost Classif...

引用

2025 IEEE/SICE International Symposium on System Integration, SII 2025

作者： Lim, Na-Young Lee, Jeon-Hyeok Park, Tae-Hyoung Chungbuk National University Department of Intelligent Systems & Robotics Cheongju Korea Republic of Chungbuk National University Department of Control & Robot Engineering Cheongju Korea Republic of

ISBN: (纸本)9798331531614

In recent developments, autonomous racing has garnered attention as it aims to overcome the limitations of standard autonomous driving systems. Achieving safe racing conditions necessitates both fast and long-range perception. However, current 3D LiDAR object detection methods face challenges with high computational costs and limited detection ranges. These issues make them unsuitable for racing scenarios. To address these challenges, this paper proposes a clustering-based long-range vehicle detection method that relies solely on LiDAR. First, the method removes ground points and foreground points and clusters the remaining points. Subsequently, these clusters are classified as vehicles using AdaBoost, generating 2D bounding boxes in the range view. Experimental results demonstrate superior performance, achieving a computational efficiency of 53 Hz and a long-range detection accuracy of over 80%, compared to voxel-based and range-based methods. This approach offers a viable solution for autonomous racing environments. © 2025 IEEE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning-Based Decision Making of Lane Change Considering Rear Vehicle Deceleration

引用

Journal of Institute of control, robotics and Systems 2022年第6期28卷 602-607页

作者： Jo, Gi-Hwan Park, Tae-Hyoung Department of Control Robot Engineering Chungbuk National University Korea Republic of Department of Intelligent Systems and Robotics Chungbuk National University Korea Republic of

This paper proposes a lane change judgment method based on deep reinforcement learning considering rear vehicle deceleration. In existing autonomous driving, the lane change problem focuses on maintaining speed or avoiding collisions. However, an accident may occur if the deceleration of the vehicle behind is not considered when changing lanes. In this paper, the network was designed to solve the lane change problem by applying the DQN (Deep Q-Network) and the Deep Sets network. Moreover, the compensation was designed based on the maintained speed and deceleration of the rear vehicle. A four-lane vehicle-only road environment was implemented as a simulation and compared with the existing method. The proposed method maintained speed and lowered the deceleration rate of the rear vehicle compared with the existing method. © ICROS 2022.

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：