Robotics and Autonomous Systems最新文献

Modeling and control of a Multirotor UAV with a cable-suspended rigid body payload: A double-pendulum approach 悬索刚体载荷多旋翼无人机建模与控制：双摆方法

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-05-26 DOI: 10.1016/j.robot.2025.105060

Mohamed Tolba, Bijan Shirinzadeh

{"title":"Modeling and control of a Multirotor UAV with a cable-suspended rigid body payload: A double-pendulum approach","authors":"Mohamed Tolba, Bijan Shirinzadeh","doi":"10.1016/j.robot.2025.105060","DOIUrl":"10.1016/j.robot.2025.105060","url":null,"abstract":"<div><div>The use of Multirotor Unmanned Aerial Vehicles (MUAVs) for aerial payload delivery has recently been a prominent research topic in the robotics community with various applications in many sectors. This paper presents a comprehensive study on the modeling and control of a generic MUAV carrier with a cable-suspended rigid body payload. A new nonlinear fifteen-degree-of-freedom dynamics model is derived accounting for practical imperfections such as shifted centers of gravity and arbitrary attachment points of the suspension cable on the carrier and payload. Additionally, this model incorporates the elastic behavior of the cable, including slackening and tightening dynamics, and introduces a double-pendulum representation for the payload. An optimal Linear Quadratic Regulator (LQR) with output weighting is proposed to stabilize the system and enable smooth tracking of the desired MUAV trajectory, while effectively reducing payload oscillations and radial movements. The developed controller successfully maintained the oscillation peak of the suspended payload below <span><math><mrow><mn>15</mn><mo>°</mo></mrow></math></span> for both stability and missions’ tracking. Unlike previous controllers, the proposed strategy regulates an arbitrary point on the MUAV, specifically the origin of its body axes, rather than the MUAV’s center of gravity or the payload. Computational analysis is used to validate the effectiveness of the proposed control strategy. The findings of this study provide valuable insights for the development of advanced MUAV-based aerial load transportation systems.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"193 ","pages":"Article 105060"},"PeriodicalIF":4.3,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhanced dynamic visual SLAM system for hospital logistics robots: Nonlinear optimal filtering, deep learning, and real-time positioning 用于医院物流机器人的增强动态视觉SLAM系统：非线性最优滤波、深度学习和实时定位

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-05-25 DOI: 10.1016/j.robot.2025.105081

Feng Xiao , Jie Fang , Xing Guo , Youhai Zhang , Rubing Huang

{"title":"Enhanced dynamic visual SLAM system for hospital logistics robots: Nonlinear optimal filtering, deep learning, and real-time positioning","authors":"Feng Xiao , Jie Fang , Xing Guo , Youhai Zhang , Rubing Huang","doi":"10.1016/j.robot.2025.105081","DOIUrl":"10.1016/j.robot.2025.105081","url":null,"abstract":"<div><div>Current Simultaneous Localization and Mapping (SLAM) systems frequently exhibit increased positioning errors, inaccuracies in map creation, and struggle with real-time multimodal sensor data fusion in dynamic environments. This paper introduces an improved SLAM system for hospital logistics robots that utilizes nonlinear optimal filtering and deep learning to navigate the challenges presented by dynamic environments. The system incorporates an Unscented Kalman Filter (UKF) for nonlinear state estimation and employs Convolutional Neural Networks (CNN) for deep feature extraction of environmental images. Semantic edge detection is accomplished through the integration of Fully Convolutional Networks (FCN) and Canny edge detection. The fusion of multimodal data is optimized using an Extended Kalman Filter (EKF) to enhance positioning accuracy across vision, lidar, and inertial measurement unit (IMU) sensors. Real-time motion estimation is achieved via an event-based camera paired with an optical flow algorithm. The proposed system delivers an absolute trajectory error (ATE) as low as 0.067 m, over 90 % overlap in map construction, and an average frame processing time under 90 ms, significantly surpassing the performance of other mainstream SLAM systems. These enhancements markedly improve positioning accuracy, mapping quality, and real-time performance in complex hospital environments.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"193 ","pages":"Article 105081"},"PeriodicalIF":4.3,"publicationDate":"2025-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Deep networks for few-shot manipulation learning from scratch 从零开始进行少量操作学习的深度网络

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-05-24 DOI: 10.1016/j.robot.2025.105056

Yinghan Chen , Xueyang Yao , Bryan Tripp

{"title":"Deep networks for few-shot manipulation learning from scratch","authors":"Yinghan Chen , Xueyang Yao , Bryan Tripp","doi":"10.1016/j.robot.2025.105056","DOIUrl":"10.1016/j.robot.2025.105056","url":null,"abstract":"<div><div>Deep networks can learn to process raw sensor data and produce control output for diverse tasks. However, to leverage these models’ flexibility and expressive power, past studies have trained them on massive amounts of data. In contrast, in this work, we attempt to train deep networks from scratch with very small datasets of object pose and gripper trajectories in manipulation-task demonstrations. The same setting has previously been used in programming-by-demonstration work with specialized statistical models such as task-parameterized Gaussian mixture models (TP-GMMs). We show that deep networks can learn manipulation tasks with performance that meets or exceeds that of past statistical models, given the same small numbers of demonstrations (5-30 in our tests), without any pretraining. Data augmentation is important for good performance and training the deep networks to be equivariant to frame transformations. Transformers performed slightly better than parameter-matched long-short-term-memory (LSTM) networks, and transformers had better training and inference times. In addition to testing these methods with physical tasks, we used a family of synthetic tasks to show that larger transformer models exhibit positive transfer across dozens of tasks, performing better on each task as they are trained on others. These results suggest that deep networks are potential alternatives to TP-GMM and related methods, having the advantage of needing fewer examples per task as the number of tasks grows. The results also suggest that the large data requirements of end-to-end manipulation learning are mainly due to perceptual factors, which may help to improve the design of end-to-end systems in the future.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"193 ","pages":"Article 105056"},"PeriodicalIF":4.3,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A wearable exoskeletal lumbar spinal rehabilitation robot based on sliding mode control scheme 一种基于滑模控制方案的可穿戴外骨骼腰椎康复机器人

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-05-23 DOI: 10.1016/j.robot.2025.105063

Chao Hou , Hongbo Wang , Fei Liu

{"title":"A wearable exoskeletal lumbar spinal rehabilitation robot based on sliding mode control scheme","authors":"Chao Hou , Hongbo Wang , Fei Liu","doi":"10.1016/j.robot.2025.105063","DOIUrl":"10.1016/j.robot.2025.105063","url":null,"abstract":"<div><div>Low back pain has become the leading cause of nonfatal health damage in the world, and about 80% of people are affected at some stage of their lives. Lumbar disc herniation (LDH) is the most common clinical cause of low back pain, of which nonsurgical treatment of lumbar traction combined with long-term bed rest can cure 80% to 90% of patients clinically. However, long-term bed rest is difficult to achieve for many people, and the use of traction equipment is expensive and inconvenient. In this paper, we developed a wearable exoskeletal lumbar rehabilitation robot (WELRR) to replace the traditional treatment of mechanical traction and long-term bed rest. The WELRR allows people to continuously wear at work and life with lumbar vertebra supported. In addition, for patients that require 2-3 months of rehabilitation after lumbar disc surgery, the robot can be used to help maintain the stability of the lumbar spine, gradually increase the range of motion of the lumbar spine in patients, and avoid muscle atrophy. The WELRR adopts the 4UPS+PS parallel bionic mechanism and sliding mode control scheme based on the linear extended observer. We verified the usability of WELRR through two experiments. In the first experiment, the maximum range of waist movement allowed by the WELRR was verified, which can meet the freedom of movement required by the human body in daily life. In the second experiment, by gathering and comparing the EMG signal of the erector spinae muscle, the use of the WELEE could reduce the EMG signal amplitude. The EMG signal while wearing the WELRR under vertical standing reached the same EMG signal level in a relaxed state, and the effectiveness of lumbar spinal decompression was verified. At present, the WELRR system is undergoing changes and testing prior to clinical trials.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"193 ","pages":"Article 105063"},"PeriodicalIF":4.3,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144178458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Path-following control using spiking neural networks associative maps 基于脉冲神经网络关联映射的路径跟踪控制

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-05-23 DOI: 10.1016/j.robot.2025.105077

Javier Pérez Fernández , Manuel Alcázar Vargas , Juan A․Cabrera Carrillo , Juan J․Castillo Aguilar , Barys Shyrokau

{"title":"Path-following control using spiking neural networks associative maps","authors":"Javier Pérez Fernández , Manuel Alcázar Vargas , Juan A․Cabrera Carrillo , Juan J․Castillo Aguilar , Barys Shyrokau","doi":"10.1016/j.robot.2025.105077","DOIUrl":"10.1016/j.robot.2025.105077","url":null,"abstract":"<div><div>Bio-inspired control systems attract significant interest in the scientific community. The advantage of neural systems lies in their ability to adapt to control processes. Path-following tasks in automated vehicles and advanced driver assistance systems are an essential component related to vehicle safety and performance. It is known that model-based controllers, which integrate a vehicle model into the control logic, are more effective than geometry-based controllers. However, a disadvantage of model-based controllers is the lack of adaptation capability to changing vehicle dynamic conditions. To address this issue, an adaptive neural controller for path-following tasks is proposed based on neural networks, particularly Spiking Neural Networks and Associative Maps. Consequently, associative maps and neural interpolation via the modelling of non-linear synaptic connections are brought to a spiking neural network to perform adaptive control tasks. Neural associative maps are used to derive functional relationships between neural inputs and outputs, further enhancing inference capabilities. In addition, neural interpolation with non-linear synaptic connections enables efficient pairwise association. Thus, by reproducing a linear quadratic regulator with a learning-capable neural network, it is possible to adjust for discrepancies and changes in dynamics through spike-timing-dependent plasticity. Results demonstrate that the adaptive controller is effective in maintaining the initial tracking performance of the vehicle while adapting to changing dynamic conditions with a computational cost that allows real-time execution. The proposed strategy results in lower error levels in lateral tracking after the learning process, while providing similar performance on heading.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"193 ","pages":"Article 105077"},"PeriodicalIF":4.3,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166101","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Saturation-tolerant prescribed performance neural formation control for air-floating robots under false data injection attacks 假数据注入攻击下浮气机器人容饱和规定性能神经编队控制

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-05-22 DOI: 10.1016/j.robot.2025.105044

Weilun Zhang , Ruiheng Hu , Guan Wang , Hongwei Xia , Guangcheng Ma

引用次数: 0

LiRA: Light-Robust Adversary for model-based reinforcement learning in real world LiRA：现实世界中基于模型的强化学习的轻鲁棒对手

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-05-22 DOI: 10.1016/j.robot.2025.105057

Taisuke Kobayashi

{"title":"LiRA: Light-Robust Adversary for model-based reinforcement learning in real world","authors":"Taisuke Kobayashi","doi":"10.1016/j.robot.2025.105057","DOIUrl":"10.1016/j.robot.2025.105057","url":null,"abstract":"<div><div>Model-based reinforcement learning has attracted much attention due to its high sample efficiency and is expected to be applied to real-world robotic applications. In the real world, as unobservable disturbances can lead to unexpected situations, robot policies should be taken to improve not only control performance but also robustness. Adversarial learning is an effective way to improve robustness, but excessive adversary would increase the risk of malfunction, and make the control performance too conservative. Therefore, this study addresses a new adversarial learning framework to make reinforcement learning robust moderately and not conservative too much. To this end, the adversarial learning is first rederived with variational inference. In addition, <em>light robustness</em>, which allows for maximizing robustness within an acceptable performance degradation, is utilized as a constraint. As a result, the proposed framework, so-called LiRA, can automatically adjust adversary level, balancing robustness and conservativeness. The expected behaviors of LiRA are confirmed in numerical simulations. In addition, LiRA succeeds in learning a force-reactive gait control of a quadrupedal robot only with real-world data collected less than two hours.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"192 ","pages":"Article 105057"},"PeriodicalIF":4.3,"publicationDate":"2025-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144138241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing constant force tracking in uncertain contact surfaces: An admittance controller utilizing virtual delayed resonator 增强不确定接触面恒力跟踪：利用虚拟延迟谐振器的导纳控制器

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-05-22 DOI: 10.1016/j.robot.2025.105008

Gang Wang , Nanzhi Xie , Honglei Che , Qi Zhang

{"title":"Enhancing constant force tracking in uncertain contact surfaces: An admittance controller utilizing virtual delayed resonator","authors":"Gang Wang , Nanzhi Xie , Honglei Che , Qi Zhang","doi":"10.1016/j.robot.2025.105008","DOIUrl":"10.1016/j.robot.2025.105008","url":null,"abstract":"<div><div>The lack of precise environmental stiffness information can significantly compromise the accuracy of force control during manipulator contact operations, leading to undesirable jitter effects at the end-effector. To address this issue, a novel admittance control method is proposed in this paper, which employs a virtual delayed resonator to enhance force tracking accuracy and suppress robot jitter. The Takagi–Sugeno (T–S) fuzzy model is developed to mitigate the impact of environmental stiffness errors, as observed by the Extended Kalman Filter (EKF), on system performance. Additionally, the <span><math><msub><mrow><mi>H</mi></mrow><mrow><mi>∞</mi></mrow></msub></math></span> control strategy, based on Linear Matrix Inequality (LMI), is implemented to ensure system stability under external disturbances. Simulation results demonstrate the effectiveness of the proposed method in achieving consistent force tracking and suppressing chatter. Experimental outcomes further validate that, even when the workpiece stiffness is unknown, the proposed approach effectively reduces jitter at the manipulator’s end-effector while improving force tracking accuracy.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"192 ","pages":"Article 105008"},"PeriodicalIF":4.3,"publicationDate":"2025-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144116842","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Tuning responsivity-persistence trade-off in swarm robotics: A motion salience threshold approach 群体机器人中响应性-持久性权衡的调整：一种运动显著性阈值方法

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-05-20 DOI: 10.1016/j.robot.2025.105055

Yanan Li , Zhicheng Zheng , Yalun Xiang , Xiaokang Lei , Xingguang Peng

{"title":"Tuning responsivity-persistence trade-off in swarm robotics: A motion salience threshold approach","authors":"Yanan Li , Zhicheng Zheng , Yalun Xiang , Xiaokang Lei , Xingguang Peng","doi":"10.1016/j.robot.2025.105055","DOIUrl":"10.1016/j.robot.2025.105055","url":null,"abstract":"<div><div>In swarm robotics, balancing two crucial properties is essential: responsivity, which enables quick reactions to environmental changes, and persistence, which maintains stable goal-directed behavior despite distractions. Responsivity is necessary for tasks like evading obstacles or responding to threats. In contrast, persistence is key to ensuring coordinated movement and focus on long-term goals, such as migration or search missions. To address the challenge of balancing these conflicting properties, we introduce the Motion Salience Threshold (MST). This approach enables swarm robots to selectively respond to significant motion cues, thereby enhancing overall system performance by minimizing unnecessary reactions to less critical changes. This tuning mechanism is particularly useful in real-world applications where the environment is unpredictable and demands both flexibility and stability from the robotic swarm. Our research demonstrates that lower threshold values increase responsivity, enabling the swarm to react quickly in highly dynamic environments, whereas higher values bolster persistence, reducing the impact of false positive signals and maintaining focus on long-term goals. The proposed approach offers a targeted method for fine-tuning swarm behavior, validated through extensive simulations and real-world experiments with robotic systems. These findings provide valuable insights for designing adaptive robotic swarms that can navigate complex and unpredictable environments more effectively.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"192 ","pages":"Article 105055"},"PeriodicalIF":4.3,"publicationDate":"2025-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144105680","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Energy-efficient motion planning for robotic systems using polynomials in the Chebyshev basis 基于切比雪夫基多项式的机器人系统节能运动规划

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-05-19 DOI: 10.1016/j.robot.2025.105051

Robbe De Laet , Nick Van Oosterwyck , Lorenzo Scalera , Annie Cuyt , Alessandro Gasparetto , Stijn Derammelaere

{"title":"Energy-efficient motion planning for robotic systems using polynomials in the Chebyshev basis","authors":"Robbe De Laet , Nick Van Oosterwyck , Lorenzo Scalera , Annie Cuyt , Alessandro Gasparetto , Stijn Derammelaere","doi":"10.1016/j.robot.2025.105051","DOIUrl":"10.1016/j.robot.2025.105051","url":null,"abstract":"<div><div>Motion profile optimization is a powerful technique for enhancing the efficiency of robotic systems without necessitating hardware modifications. Nonetheless, the prevailing usage of piecewise or polynomial position functions can often require a high number of design parameters or result in unbounded optimization problems. This paper presents a novel approach employing polynomials expressed in the Chebyshev basis for the position function of multi-degree-of-freedom (DOF) systems, enabling substantial performance improvements with a minimal number of design parameters while enabling the use of a bounded design space. More specifically, this work focuses on reducing energy consumption while maintaining a fixed motion time. Moreover, by symbolically formulating the motion profile, it is demonstrated that kinematic constraints can be linearized, leading to accelerated convergence in the optimization process. To illustrate the robustness of the proposed method under different operational conditions, optimizations were executed on three distinct motion tasks and a range of payload values, and compared to a state-of-the-art method. Experimental results strongly validate the effectiveness of the proposed approach, demonstrating a reduction in root mean square (rms) torque by up to −47.6% with a limited number of design parameters for each joint.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"192 ","pages":"Article 105051"},"PeriodicalIF":4.3,"publicationDate":"2025-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144130792","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0