Robotics and Autonomous Systems最新文献

{"title":"Virtual framework customizing wearable mechanisms: An application to gait-assistive mobile hip robots","authors":"Jegyeong Ryu ,&nbsp;Jongjun Lee ,&nbsp;Jeonghan Yu ,&nbsp;Seok Won Kang ,&nbsp;Yoon Young Kim","doi":"10.1016/j.robot.2025.105052","DOIUrl":"10.1016/j.robot.2025.105052","url":null,"abstract":"<div><div>Wearable robots are now beginning to transition into consumer products for everyday use, moving beyond their traditional industrial applications. For such applications, personalization is crucial to maximize the effectiveness of the robots while avoiding discomfort. We propose a virtual computational design framework specifically developed to personalize linkage mechanisms in wearable robots. This framework focuses on the design of force-transmission capabilities for gait-assistive hip wearable robot mechanisms. Incorporating an individual’s musculoskeletal modeling and gait motion, it optimizes the mechanism shape, which governs the relationship between the wearer’s hip angle and the assistive moment direction. Also, the framework simultaneously updates the actuator torque profile to maximally reduce the metabolic cost evaluated through musculoskeletal analysis. The proposed framework's effectiveness is demonstrated by customizing a single-actuator hip wearable robot; for several individuals considered, the metabolic costs for the customized robots were reduced compared to those for robots before customization or partially customized ones where only the control profiles were considered. We expect the proposed customization framework to improve the usability of consumer wearable robots by offering an affordable and lightweight personalized robot through a user-friendly virtual platform with minimal effort required from the consumer.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"192 ","pages":"Article 105052"},"PeriodicalIF":4.3,"publicationDate":"2025-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143924220","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}

{"title":"Full coverage path planning strategy for cleaning robots in semi-structured outdoor environments","authors":"Kai Wu ,&nbsp;Zihao Wu ,&nbsp;Shaofeng Lu ,&nbsp;Weihua Li","doi":"10.1016/j.robot.2025.105050","DOIUrl":"10.1016/j.robot.2025.105050","url":null,"abstract":"<div><div>With the increasing demand for autonomous cleaning solutions in urban environments, Coverage Path Planning (CPP) technology has seen widespread development in the field of cleaning robots. Cleaning robots must navigate complex semi-structured environments, which are often characterized by intricate area layouts and diverse obstacles. This study proposes a CPP framework for semi-structured environments, integrating a map preprocessor and a coverage path planner. The map preprocessor accurately processes environmental boundaries and obstacles, optimizing the environmental information to enable the coverage path planner to generate more efficient paths. This improves coverage efficiency, allowing cleaning robots to adapt to semi-structured community environments. Both simulation and real-world experiments demonstrate that the proposed framework offers significant performance advantages over state-of-the-art methods. By maximizing the coverage rate and minimizing the number of turns in the coverage path, our approach significantly enhances autonomous cleaning robots' cleaning efficiency and quality. This research improves the operational capabilities of cleaning robots in semi-structured environments and provides valuable insights and practical guidance for the broader field of autonomous path planning.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"192 ","pages":"Article 105050"},"PeriodicalIF":4.3,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143912660","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}

{"title":"Curriculum-guided skill learning for long-horizon robot manipulation tasks","authors":"João Bernardo Alves,&nbsp;Nuno Lau,&nbsp;Filipe Silva","doi":"10.1016/j.robot.2025.105032","DOIUrl":"10.1016/j.robot.2025.105032","url":null,"abstract":"<div><div>Robotic tasks often involve solving long-horizon problems. Seen under the reinforcement learning framework, the rewards provided in these problems are often sparse, which can be problematic for the learning process. In this context, dividing the long-horizon task into smaller ones represents a viable strategy to alleviate the credit assignment problem. Another approach generally used to help with this problem is curriculum learning. This paper combines both with a new skill-chaining learning algorithm that provides transition policies to bridge the gap between skills. Our approach begins by extracting meaningful skills from the states of an expert trajectory, using a heuristic method, which are subsequently used by the skill learning and the skill chaining algorithms. By leveraging the sequential order of the skills inside the demonstration, we propose a method to learn inter-skill transition policies to ensure the skills are appropriately chained. Our curriculum-based training approach enables an agent to learn action sequences that generalize inside a specific sub-task context. Using the information of a single demonstration, we show that our approach can solve a robotic manipulation task with similar performance to methods that rely on a large amount of data. Because our skill segmentation method detects which skills are present across demonstrations, we also show that our approach can reuse skills already learned in a zero-shot way.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"192 ","pages":"Article 105032"},"PeriodicalIF":4.3,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143912661","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}

{"title":"Model Predictive Control-based dynamic movement primitives for trajectory learning and obstacle avoidance","authors":"Tao Ma,&nbsp;Burkhard Corves","doi":"10.1016/j.robot.2025.105027","DOIUrl":"10.1016/j.robot.2025.105027","url":null,"abstract":"<div><div>Dynamic Movement Primitives (DMP) is a well-established framework in Learning from Demonstration, widely used for acquiring and reproducing motion patterns. To improve DMP’s adaptability and generalization in complex environment, this paper introduces a novel DMP obstacle avoidance approach by integrating Model Predictive Control (MPC). The proposed method formalizes DMP within an MPC framework and employs the Proximal Averaged Newton for Optimal Control solver to incorporate obstacle avoidance conditions into the cost function, adapting to various environments and improving obstacle avoidance efficiency. Evaluations were conducted in scenarios involving static and dynamic obstacles across two- and three-dimensional spaces, as well as multi-obstacle and dynamic environments. It adapts to the scaling characteristics of DMP, demonstrating robustness in complex scenarios. Additionally, the inclusion of a safety threshold enhances the method’s robustness, ensuring safe obstacle avoidance in uncertain environments. Quantitative comparisons with Artificial Potential Field and Steering Angle approaches highlight superior performance in trajectory efficiency and obstacle avoidance success. Finally, validation through simulations and real-world experiments, including a pick-and-place task on UR5 robot, demonstrate the approach’s practical applicability in robotic operations.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"192 ","pages":"Article 105027"},"PeriodicalIF":4.3,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143904120","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}

{"title":"Distributed cooperative pursuit with encirclement guarantee via robust model predictive control","authors":"Chen Wang ,&nbsp;Hua Chen ,&nbsp;Jia Pan ,&nbsp;Wei Zhang","doi":"10.1016/j.robot.2025.105019","DOIUrl":"10.1016/j.robot.2025.105019","url":null,"abstract":"<div><div>This paper studies a cooperative pursuit-evasion problem with the encirclement guarantee. Inspired by the importance of forming an encirclement of the evader in practical applications such as surveillance, we explicitly consider the encirclement condition throughout the game in the synthesis of a cooperative pursuit strategy. Different from classical pursuit-evasion problems, the studied problem requires a careful balance between the capture and encirclement conditions. To solve this challenging problem, we exploit the geometric nature of the encirclement condition and develop a robust model predictive control (RMPC) based strategy to account for the encirclement and capture requirements jointly. To address the bilinearity of the centralized RMPC problem and to respect the communication constraint, we proposed a novel approach to decouple the pursuit-evasion problem among the pursuers. Such a distributed strategy relies on a novel Encirclement-Guaranteed Sectors Set (EGSS) concept that inner-approximates the original bilinear RMPC problem with a set of decoupled linear RMPC problems. These linear problems can be efficiently solved by Tube Model Predictive Control (TMPC) with only local information. To validate the effectiveness of the proposed framework, we provide extensive simulation experiments with realistic mobile robot models. Comparisons with the baseline Voronoi-based strategy demonstrate the robustness of the proposed approach in guaranteeing encirclement while achieving successful capture.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"192 ","pages":"Article 105019"},"PeriodicalIF":4.3,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144069116","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}

{"title":"Estimation of aerodynamic effects in coaxial rotors for UAVs using fractional Unscented Kalman Filter and DCSV","authors":"C.A. Peña Fernández ,&nbsp;Florian Holzapfel","doi":"10.1016/j.robot.2025.105028","DOIUrl":"10.1016/j.robot.2025.105028","url":null,"abstract":"<div><div>Coaxial rotors, especially in larger Unmanned Aerial Vehicles (UAVs), introduce new challenges in aerodynamic modeling and control due to complex rotor interactions. Known for their efficiency in compact designs, coaxial rotors face issues such as rotor downwash interference, turbulence, and reduced aerodynamic efficiency, all of which can severely impact flight stability and control precision. Traditional Kalman filters (KF) are often an efficient alternative to address these problems, but they are insufficient for the nonlinear dynamics of UAVs, leading to the adoption of the Unscented Kalman Filter (UKF) for more precise state estimation. However, UKF’s performance is limited by the non-local nature of aerodynamic effects. This paper proposes a two-step method to address these challenges. First, by incorporating fractional-order derivatives, the Merwe point selection technique, and the DCSV (Dynamic Confined Spaces of Velocities) criterion for multiple time scales, the proposed FUKF enhances real-time prediction accuracy. Second, to demonstrate the efficiency of the signals observed by the FUKF, a weak sparse identification technique for nonlinear dynamics with weighted integration and the Bayesian MCMC (Monte Carlo Markov Chain) method are used offline to identify aerodynamic equations, accounting for rotor-rotor interference and complex airflows. The results show that for different time scales, classical KF models fail to accurately predict the states associated with aerodynamic effects due to the lack of inclusion of parametric interdependence, a concept introduced in a forthcoming lemma. This approach improves the modeling accuracy of aerodynamic effects for coaxial rotors and represents an advancement in the performance of estimation techniques for UAVs.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"191 ","pages":"Article 105028"},"PeriodicalIF":4.3,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143890901","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}

{"title":"ASTRA-BF: A human-in-the-loop algorithm for predicting surface transitions in robotic lower limb prosthetics using biomechanical features","authors":"Charikleia Angelidou ,&nbsp;Jaclyn M. Sions ,&nbsp;Panagiotis Artemiadis","doi":"10.1016/j.robot.2025.105030","DOIUrl":"10.1016/j.robot.2025.105030","url":null,"abstract":"<div><div>In the realm of lower-limb prosthetics and wearable devices, the significance of walking on compliant surfaces for individuals with lower-limb amputations (LLA) cannot be overstated. The interaction between humans and compliant surfaces presents a unique challenge, particularly for those relying on prosthetic interventions. Ensuring the safety, stability, and fluidity of movement on these surfaces is paramount for preventing falls in this population. This work delves into the critical importance of addressing this challenge, outlining the complexities involved in walking on compliant surfaces, and exploring high-level control strategy interventions aimed at mitigating the inherent difficulties faced by individuals with LLA. We specifically expand on an individualized pattern recognition (PR) and classification approach utilizing kinematic, kinetic, and surface electromyography (EMG) data to discern user intent for transitioning from rigid to compliant surfaces of variable stiffness. This work proposes an efficient Algorithm for Surface TRAnsition prediction in robotic lower limb prosthetics using Biomechanical Features, called ASTRA-BF. Integrating a k-Nearest Neighbors (kNN) methodology alongside a Naive Bayes classification, our strategy can accurately forecast impending transitions in four different levels of surface stiffness in real time. This capability enables swift parameter control of prostheses or wearable devices, facilitating adaptation to diverse terrains. Post-implementation of the ASTRA-BF, classification results attain a prediction accuracy of up to 82%, demonstrating the feasibility and efficiency of real-time prediction for transitions to various compliant surfaces in both healthy and clinical populations. The proposed framework has the potential to propel the field of robotics toward novel solutions that not only enhance mobility but also significantly improve the quality of life for those navigating the intricate terrain of compliant surfaces with prosthetic limbs.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"191 ","pages":"Article 105030"},"PeriodicalIF":4.3,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143894731","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}

{"title":"Robot localization aided by quantum algorithms","authors":"Unai Antero ,&nbsp;Basilio Sierra ,&nbsp;Jon Oñativia ,&nbsp;Alejandra Ruiz ,&nbsp;Eneko Osaba","doi":"10.1016/j.robot.2025.105026","DOIUrl":"10.1016/j.robot.2025.105026","url":null,"abstract":"<div><div>Localization is a critical aspect of mobile robotics, enabling robots to navigate their environment efficiently and avoid obstacles.</div><div>Current probabilistic localization methods, such as the Adaptive Monte Carlo localization (AMCL) algorithm, are computationally intensive and may struggle with large maps or high resolution sensor data.</div><div>This paper explores the application of quantum computing in robotics, focusing on the use of Grover’s search algorithm to improve the efficiency of localization in mobile robots. We propose a novel approach to utilize Grover’s algorithm in a 2D map, enabling faster and more efficient localization.</div><div>Despite the limitations of current physical quantum computers, our experimental results demonstrate a significant speedup over classical methods, highlighting the potential of quantum computing to improve robotic localization. This work bridges the gap between quantum computing and robotics, providing a practical solution for robotic localization and paving the way for future research in quantum robotics.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"192 ","pages":"Article 105026"},"PeriodicalIF":4.3,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143900251","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}

{"title":"From data extraction to data-driven dynamic modeling for cobots: A method using multi-objective optimization","authors":"Diego Navarro-Cabrera ,&nbsp;Juan H. García-Guzmán ,&nbsp;Nicolás C. Cruz ,&nbsp;Brayan Valencia-Vidal ,&nbsp;Niceto R. Luque ,&nbsp;Eduardo Ros","doi":"10.1016/j.robot.2025.105006","DOIUrl":"10.1016/j.robot.2025.105006","url":null,"abstract":"<div><div>Controlling collaborative robots (cobots) is a new and challenging paradigm within the field of robot motion control and safe human–robot interaction (HRI). The safety measures needed for a reliable interaction between the robot and its environment hinder the use of classical position control methods, pushing researchers to explore alternative motor control techniques, with a strong focus on those rooted in machine learning (ML). While reinforcement learning has emerged as the predominant approach for creating intelligent controllers for cobots, supervised learning represents a promising alternative in developing data-driven model-based ML controllers in a faster and safer way. In this work, we study several aspects of the methodology needed to create a dataset for learning the dynamics of a robot. To this aim, we fine-tune several PD controllers across different benchmark trajectories using multi-objective evolutionary algorithms (MOEAs) that take into account controller accuracy, and compliance in terms of low torques in the framework of safe HRI. We delve into various aspects of the data extraction methodology including the selection and calibration of the MOEAs. We also demonstrate the need to tune controllers individually for each trajectory and how the speed of a trajectory influences both the tuning process and the resulting dynamics of the robot. Finally, we create a novel dataset and validate its use by feeding all the extracted dynamic data into an inverse dynamic robot model and integrating it into a feedforward control loop. Our approach significantly outperforms individual standard PD controllers previously tuned, thus illustrating the effectiveness of the proposed methodology.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"191 ","pages":"Article 105006"},"PeriodicalIF":4.3,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143890902","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}

{"title":"RUMOR: Reinforcement learning for understanding a model of the real world for navigation in dynamic environments","authors":"Diego Martinez-Baselga,&nbsp;Luis Riazuelo,&nbsp;Luis Montano","doi":"10.1016/j.robot.2025.105020","DOIUrl":"10.1016/j.robot.2025.105020","url":null,"abstract":"<div><div>Autonomous navigation in dynamic environments is a complex but essential task for autonomous robots, with recent deep reinforcement learning approaches showing promising results. However, the complexity of the real world makes it infeasible to train agents in every possible scenario configuration. Moreover, existing methods typically overlook factors such as robot kinodynamic constraints, or assume perfect knowledge of the environment. In this work, we present RUMOR, a novel planner for differential-drive robots that uses deep reinforcement learning to navigate in highly dynamic environments. Unlike other end-to-end DRL planners, it uses a descriptive robocentric velocity space model to extract the dynamic environment information, enhancing training effectiveness and scenario interpretation. Additionally, we propose an action space that inherently considers robot kinodynamics and train it in a simulator that reproduces the real world problematic aspects, reducing the gap between the reality and simulation. We extensively compare RUMOR with other state-of-the-art approaches, demonstrating a better performance, and provide a detailed analysis of the results. Finally, we validate RUMOR’s performance in real-world settings by deploying it on a ground robot. Our experiments, conducted in crowded scenarios and unseen environments, confirm the algorithm’s robustness and transferability.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"191 ","pages":"Article 105020"},"PeriodicalIF":4.3,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143887414","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}