Robotics and Autonomous Systems最新文献

Hierarchical residual reinforcement learning based path tracking control method for unmanned bicycle

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-03-31 DOI: 10.1016/j.robot.2025.104996

Benyan Huo , Long Yu , Yanhong Liu , Zhang Chen

{"title":"Hierarchical residual reinforcement learning based path tracking control method for unmanned bicycle","authors":"Benyan Huo , Long Yu , Yanhong Liu , Zhang Chen","doi":"10.1016/j.robot.2025.104996","DOIUrl":"10.1016/j.robot.2025.104996","url":null,"abstract":"<div><div>Due to their super under-actuation and high nonlinearity properties, simplification and linearization techniques are necessary in the traditional model-based path tracking controller design of unmanned bicycles, leading to the decrease of their control performance. On the other hand, as one of the most important learning-based methods, deep reinforcement learning (DRL) suffers from low training efficiency and initial safety issues. In this letter, a hierarchical residual reinforcement learning (HRRL)-based path tracking control method is proposed, to address the drawbacks of both traditional and learning-based approaches. The path tracking task is decomposed into two subtasks, i.e., attitude control and position control, and the controllers are designed separately for each subtask. In each controller, a DRL controller is connected to a traditional controller through residual connection. Physical simulation experiments demonstrate that compared to the traditional LQR, LQI, Stanley and DRL approaches, the proposed method can improve the tracking performance of unmanned bicycles and decrease the training time and the tipping number during training. Furthermore, experimental results also show that the proposed controller exhibits a certain level of robustness, enabling effective path tracking in complex terrain.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"190 ","pages":"Article 104996"},"PeriodicalIF":4.3,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143759962","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

Fall analysis and prediction for humanoids

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-03-29 DOI: 10.1016/j.robot.2025.104995

Chiyu Zhang, Jie Gao, Ziyu Chen, Shanlin Zhong, Hong Qiao

{"title":"Fall analysis and prediction for humanoids","authors":"Chiyu Zhang, Jie Gao, Ziyu Chen, Shanlin Zhong, Hong Qiao","doi":"10.1016/j.robot.2025.104995","DOIUrl":"10.1016/j.robot.2025.104995","url":null,"abstract":"<div><div>In bipedal locomotion, avoiding falls is a significant and challenging issue, and addressing this challenge requires precise fall prediction methods. Since humanoid robots’ state data is a temporal signal in a high-dimensional feature space, how to extract appropriate features and enable rapid real-time response in actual systems constitutes a significant challenge. We designed an algorithm of convex hull vertexes selection (CHVS) for fall analysis, selecting features that contribute more to assessing the robot’s state. For the fall prediction, we first used an 11-dimensional statistical measure for feature extraction from the temporal signal. After state classification, we further incorporated the temporal accumulation effect, achieving a trade-off between minimizing false positive rates and maximizing advance prediction time. Subsequently, we designed a simple adjustment strategy for imminent falls and integrated it into the overall control framework. We validated the proposed fall prediction algorithm using the data of robot Q1. The validation was carried out under the condition of small sample training. We compared our algorithm with mainstream learning-based fall state classification methods, such as Long Short Term Memory Networks (LSTM) and Convolutional Neural Networks (CNN). The results show that the accuracy of state classification has increased by 3%–5%. At the same time, the computation time has been reduced by more than half. After time accumulation, the algorithm could predict more than 1 s in advance with a false positive rate of 0. The control framework incorporating the fall adjustment was tested in the simulation of the Cassie robot with an upper body. Some disturbances that could not be overcome under the original framework were adjusted to a normal motion state, proving the usefulness and feasibility of the proposed fall prediction for real-time dynamic scenarios to prevent falls.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"190 ","pages":"Article 104995"},"PeriodicalIF":4.3,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143739271","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

Magnetically actuated millimeter-scale biped

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-03-25 DOI: 10.1016/j.robot.2025.104985

Adam Cox, Farshid Asadi, Ammar Yacoub, Sinan Beskok, Yildirim Hurmuzlu

{"title":"Magnetically actuated millimeter-scale biped","authors":"Adam Cox, Farshid Asadi, Ammar Yacoub, Sinan Beskok, Yildirim Hurmuzlu","doi":"10.1016/j.robot.2025.104985","DOIUrl":"10.1016/j.robot.2025.104985","url":null,"abstract":"<div><div>This paper presents a novel approach to studying bipedal locomotion using magnetically actuated miniature robots. Traditional bipedal locomotion machines are expensive and complex. In contrast, we introduce “<em>Big Foot</em>”, a lightweight 0.3 g robot designed to explore fundamental concepts of bipedal locomotion without requiring complex hardware.</div><div>We introduce a novel design and actuation mechanism for the magnetically actuated miniature robot, followed by a numerical model to analyze its dynamics and stability. Two actuation schemes are investigated: heel strike-based and constant period actuation. Heel strike-based actuation provides superior stability, consistent gait generation, and faster locomotion, while constant period actuation excels on steeper slopes. Experimental validation with <em>Big Foot</em> confirms successful uphill gait generation and trajectory tracking. Additionally, we highlight the potential scalability of our approach, proposing an extension to a design similar to an inertially actuated larger-scale biped without joint actuation at the ankles or hips.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"190 ","pages":"Article 104985"},"PeriodicalIF":4.3,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143739270","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

High-safety path optimization for mobile robots using an improved ant colony algorithm with integrated repulsive field rules

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-03-24 DOI: 10.1016/j.robot.2025.104998

Jianjuan Liu , Yiheng Qian , Wenzhuo Zhang , Miaoxin Ji , Qiangwei Xv , Hongliang Song

{"title":"High-safety path optimization for mobile robots using an improved ant colony algorithm with integrated repulsive field rules","authors":"Jianjuan Liu , Yiheng Qian , Wenzhuo Zhang , Miaoxin Ji , Qiangwei Xv , Hongliang Song","doi":"10.1016/j.robot.2025.104998","DOIUrl":"10.1016/j.robot.2025.104998","url":null,"abstract":"<div><div>This paper introduces an improved ant colony algorithm, AR-ACO (A*-Repulsive field-ACO), to enhance the efficiency of path planning for mobile robots in complex maps. This paper adopts six strategies to optimize the iterative process of the ACO algorithm and strengthen the selection of the optimal path. Firstly, inspired by the repulsive field characteristics of artificial potential fields, an Obstacle Impact Factor is introduced to identify the risk zones in the map and adjust the initial distribution of pheromones. Secondly, to ensure the global and discriminative nature of the map, a dynamic constraint is applied to pheromones, following the MAX-MIN Ant system. Thirdly, a novel backtracking mechanism is proposed to address deadlock situations and reduce computational burden. Fourthly, heuristic information in the ant colony algorithm is improved to accelerate convergence speed and enhance the smoothness of global paths. Fifthly, the evaporation factor pheromone formula has been improved to optimize the ability to cope with complex terrains. Additionally, an improved elite ant retention strategy is introduced to significantly enhance the ants' optimization capability while ensuring convergence speed. Simulation experiments and physical verifications conducted in various environments, especially in complex large-scale maps, demonstrate that the optimized algorithm outperforms traditional algorithms, confirming the effectiveness of the improved ant colony algorithm.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"190 ","pages":"Article 104998"},"PeriodicalIF":4.3,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143735001","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

Imitation learning-based Direct Visual Servoing using the large projection formulation

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-03-20 DOI: 10.1016/j.robot.2025.104971

Sayantan Auddy , Antonio Paolillo , Justus Piater , Matteo Saveriano

引用次数: 0

Neural network-based intelligent perception guaranteed performance control for mechanical arm

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-03-18 DOI: 10.1016/j.robot.2025.104991

Chunwu Yin , Pei Yi , Xiangwei Bu

{"title":"Neural network-based intelligent perception guaranteed performance control for mechanical arm","authors":"Chunwu Yin , Pei Yi , Xiangwei Bu","doi":"10.1016/j.robot.2025.104991","DOIUrl":"10.1016/j.robot.2025.104991","url":null,"abstract":"<div><div>We concern with the intelligent perception control of a mechanical arm dynamic system with unknown initial state values, parameter perturbations, and external disturbances. Unlike existing prescribed performance control (PPC) methodologies which fail to preset convergence time via parameter setting, we propose a new type of PPC with predefined convergence time to impose prescribed behaviors on angle tracking errors. To accomplish such aim, we firstly define a predefined time stability criterion with an upper bound of convergence time that can be set in advance, and then we further convert the actual tracking error variable into a new variable with an initial value of zero by utilizing the error conversion function. Furthermore, a boundary amplitude intelligent extension algorithm is designed based on tracking error for performance constraint function (PCF), and meanwhile the radial basis function neural network (RBFNN) is adopted to approximate the mechanical arm system model. On this basis, a new PPC approach guaranteeing predefined convergence time is addressed for the mechanical arm system. Finally, the obtained simulation results reveal that the angle tracking error always evolves inside the extended boundary of the PCF, to satisfy better prescribed transient and steady-state properties in comparison with existing technics.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"190 ","pages":"Article 104991"},"PeriodicalIF":4.3,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704892","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

Optimal path planning for unmanned aerial vehicles with multiple round-trip flights in coverage tasks

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-03-15 DOI: 10.1016/j.robot.2025.104970

Jing Li , Yonghua Xiong , Jinhua She , Anjun Yu

{"title":"Optimal path planning for unmanned aerial vehicles with multiple round-trip flights in coverage tasks","authors":"Jing Li , Yonghua Xiong , Jinhua She , Anjun Yu","doi":"10.1016/j.robot.2025.104970","DOIUrl":"10.1016/j.robot.2025.104970","url":null,"abstract":"<div><div>As high-tech equipment for rescue and relief, unmanned aerial vehicles (UAVs) are widely used in remote relief operations during disasters, significantly improving the efficiency of rescue efforts. However, one significant challenge of UAVs is the limitation of their onboard battery, which prohibits them from completing coverage tasks in a single journey, requiring multiple round-trip flights and frequent battery charging or replacement. As a result, it will greatly prolong the task time. To improve the efficiency of coverage tasks, we allocate task points reasonably to minimize the coverage rounds, and carry out path planning to optimize the travel time of each UAV. This study first formulates a path planning model with the optimization objective of minimizing the overall task time. Then, a task allocation strategy is designed based on the priority of task points, including a max-weight allocation scheme for special scenarios with absolute priority rules and a min-delay allocation scheme for general scenarios with relative priority rules. To optimize the paths of UAVs, we further develop an improved beetle antennae search algorithm based on mutation operations (MBAS). The performance of the developed integrated methods is finally tested through simulation, yielding good results. Source code of the algorithm can be found at <span><span>https://github.com/lijing0966/MBAS.git</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"189 ","pages":"Article 104970"},"PeriodicalIF":4.3,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143642690","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

Towards a robotic intrusion prevention system: Combining security and safety in cognitive social robots

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-03-15 DOI: 10.1016/j.robot.2025.104959

Francisco Martín , Enrique Soriano-Salvador , José Miguel Guerrero , Gorka Guardiola Múzquiz , Juan Carlos Manzanares , Francisco J. Rodríguez

{"title":"Towards a robotic intrusion prevention system: Combining security and safety in cognitive social robots","authors":"Francisco Martín , Enrique Soriano-Salvador , José Miguel Guerrero , Gorka Guardiola Múzquiz , Juan Carlos Manzanares , Francisco J. Rodríguez","doi":"10.1016/j.robot.2025.104959","DOIUrl":"10.1016/j.robot.2025.104959","url":null,"abstract":"<div><div>Social Robots need to be safe and reliable to share their space with humans. This paper reports on the first results of a research project that aims to create more safe and reliable, intelligent autonomous robots by investigating the implications and interactions between cybersecurity and safety. We propose creating a robotic intrusion prevention system (RIPS) that follows a novel approach to detect and mitigate intrusions in cognitive social robot systems and other cyber–physical systems. The RIPS detects threats at the robotic communication level and enables mitigation of the cyber–physical threats by using <em>System Modes</em> to define what part of the robotic system reduces or limits its functionality while the system is compromised. We demonstrate the validity of our approach by applying it to a cognitive architecture running in a real social robot that preserves the privacy and safety of humans while facing several cyber attack situations.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"190 ","pages":"Article 104959"},"PeriodicalIF":4.3,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143725806","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

Resilient nonlinear model predictive control for formation-containment of multi-mobile robot systems

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-03-15 DOI: 10.1016/j.robot.2025.104983

Alireza Kazemi, Iman Sharifi

{"title":"Resilient nonlinear model predictive control for formation-containment of multi-mobile robot systems","authors":"Alireza Kazemi, Iman Sharifi","doi":"10.1016/j.robot.2025.104983","DOIUrl":"10.1016/j.robot.2025.104983","url":null,"abstract":"<div><div>This paper focuses on resilient nonlinear model predictive control (NMPC) for the formation containment of multiple nonholonomic mobile robots in the presence of Denial-of-Service (DoS) attacks. The proposed strategy addresses obstacle and collision avoidance between agents by defining a safe circular region for each agent. The scenario-based cost function of NMPC encompasses terms dedicated to achieving the desired formation by leaders, converging the states of followers to the convex hull spanned by leaders, and minimizing control efforts. Utilizing an acknowledgment-based packet transmission strategy, coupled with a buffer mechanism on the actuator side, alleviates the impact of control signal absence during DoS attacks on the controller-to-actuator (C-A) channel. As a Lyapunov-based approach, the contractive constraint in MPC is employed to establish the stability of Multi-Robot Systems (MRS) throughout the mission. A search and rescue application, utilized as a simulation case study, verifies the proposed method’s usefulness and efficiency. Moreover, In the evaluation of real-time implementation, the proposed scheme was validated through a laboratory-based experiment involving a customized mobile robot and low-cost hardware-in-the-loop (HIL) agents based on Raspberry Pi.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"189 ","pages":"Article 104983"},"PeriodicalIF":4.3,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143642689","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

Development of a soft gripper for replicating human grasps in forest nursery tasks

IF 4.3 2区计算机科学

Robotics and Autonomous Systems Pub Date : 2025-03-14 DOI: 10.1016/j.robot.2025.104987

Mohammad Sheikh Sofla , Hanita Golshanian , Elizabeth I. Sklar , Marcello Calisti

{"title":"Development of a soft gripper for replicating human grasps in forest nursery tasks","authors":"Mohammad Sheikh Sofla , Hanita Golshanian , Elizabeth I. Sklar , Marcello Calisti","doi":"10.1016/j.robot.2025.104987","DOIUrl":"10.1016/j.robot.2025.104987","url":null,"abstract":"<div><div>This research aims to automate labour-intensive tasks in forest nurseries by developing a soft gripper that mimics human workers' grasps to perform the singulation and sorting of tree saplings. By analysing human workers and conducting experimental investigations, the required grasp types and grip forces were identified. The Fin Ray Effect (FRE) structure, noted for its adaptability to asymmetric shapes, was chosen as the gripper's basis. However, modifications were necessary to achieve the required power and pinch grasp types and to provide the desired grip forces. Simulation analysis explored various beam configurations and boundary conditions of FRE fingers, resulting in a proposed modified design. Experimental investigations confirmed that the proposed gripper effectively delivered required grasps and grip forces. The new design enabled three additional grasp types for FRE grippers and increased grip forces by over 200 %. This gripper design is suitable for industrial pick-and-place applications where precise pinching grasp and various power grasps with sufficient payload capacity are needed.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"189 ","pages":"Article 104987"},"PeriodicalIF":4.3,"publicationDate":"2025-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143680837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0