Robotics and Autonomous Systems最新文献

{"title":"An integral terminal sliding mode-based adaptive control approach for traversing unknown inclined surfaces","authors":"Lin Zhang ,&nbsp;Lei Chen ,&nbsp;Muhammad Saqib ,&nbsp;Baoyu Wang ,&nbsp;Pengjie Xu ,&nbsp;Yanzheng Zhao","doi":"10.1016/j.robot.2025.104928","DOIUrl":"10.1016/j.robot.2025.104928","url":null,"abstract":"<div><div>Skid-steering control is commonly used in mobile robots, but its application to climbing manipulation-oriented robots (CMo-R) requires further development. This study proposes an adaptive skid-steering control strategy using an integral terminal sliding mode controller (ITSMC) to improve the climbing maneuverability of four-wheeled CMo-Rs on unknown inclined surfaces. The control law is developed using both kinematics and dynamics models, considering slipping effects to reduce slippage during 3D motion. A slip estimation and ITSMC-based adaptive control algorithm are introduced to enhance tracking accuracy in complex 3D environments. The proposed approach is compared to traditional PID and adaptive kinematic controllers through simulations and experiments. Results show that the proposed method outperforms the others in terms of tracking performance and robustness, especially for navigating horizontal, inclined, and vertical surfaces. This work provides a new control strategy for CMo-Rs, contributing to the feasibility and stability of future climbing manipulation applications.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"186 ","pages":"Article 104928"},"PeriodicalIF":4.3,"publicationDate":"2025-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095628","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":"Efficiency-optimized path planning algorithm for car-like mobile robots in bilateral constraint corridor environments","authors":"Junkui Zhong ,&nbsp;Deyi Kong ,&nbsp;Yuliang Wei ,&nbsp;Xiaojuan Hu ,&nbsp;Yang Yang","doi":"10.1016/j.robot.2025.104923","DOIUrl":"10.1016/j.robot.2025.104923","url":null,"abstract":"<div><div>In the field of mobile robot path planning, optimizing mobility efficiency is paramount for enhancing operational productivity. This paper presents a novel path planning algorithm designed to optimize mobility efficiency. The algorithm generates free paths and employs turning points for segmentation, while Dubins and clothoid curves are utilized for path smoothing within kinematic constraints. An evaluation function, considering dynamic variables like velocity loss and distance traveled during turning, selects the optimal path for mobility efficiency. Experimental results reveal that the shortest path in length is not always the most efficient. Comparative analysis with the Hybrid A* algorithm showcases the proposed algorithm’s ability to generate smooth paths across various constraint environments, thereby enhancing robot mobility. Validation experiments on a custom-developed three-wheeled mobile robot confirm the effectiveness of the derived paths. This efficiency-optimized path planning algorithm finds practical application in settings such as factories with dual-boundary constraints and intricate corner configurations, offering time-saving trajectories to bolster overall robot operational efficacy.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"186 ","pages":"Article 104923"},"PeriodicalIF":4.3,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095627","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":"RainSD: Rain style diversification module for image synthesis enhancement using feature-level style distribution","authors":"Hyeonjae Jeon ,&nbsp;Junghyun Seo ,&nbsp;Taesoo Kim ,&nbsp;Sungho Son ,&nbsp;Jungki Lee ,&nbsp;Gyeungho Choi ,&nbsp;Yongseob Lim","doi":"10.1016/j.robot.2025.104922","DOIUrl":"10.1016/j.robot.2025.104922","url":null,"abstract":"<div><div>Autonomous driving technology nowadays targets to level 4 or beyond, but the researchers are faced with some limitations for developing reliable driving algorithms in diverse challenges. To promote the spread of autonomous vehicles widely, it is important to address safety issues in this technology. Among various safety concerns, the sensor blockage problem by severe weather conditions can be one of the most frequent threats for multi-task learning-based perception algorithms during autonomous driving. To handle this problem, the importance of generating proper datasets is becoming more significant. In this paper, a synthetic road dataset with sensor blockage generated from real road dataset BDD100K is suggested in the format of BDD100K annotation. Rain streaks for each frame were made using an experimentally established equation and translated utilizing the image-to-image translation network based on style transfer. Using this dataset, the degradation of the diverse multitask networks for autonomous driving, such as lane detection, driving area segmentation, and traffic object detection, has been thoroughly evaluated and analyzed. The tendency of performance degradation of deep neural network-based perception systems for autonomous vehicles has been analyzed in depth. Finally, we discuss the limitation and future directions of deep neural network-based perception algorithms and autonomous driving dataset generation based on image-to-image translation.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"186 ","pages":"Article 104922"},"PeriodicalIF":4.3,"publicationDate":"2025-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095622","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":"dGrasp: NeRF-Informed implicit grasp policies with supervised optimization slopes","authors":"Gergely Sóti ,&nbsp;Xi Huang ,&nbsp;Christian Wurll ,&nbsp;Björn Hein","doi":"10.1016/j.robot.2025.104921","DOIUrl":"10.1016/j.robot.2025.104921","url":null,"abstract":"<div><div>We present dGrasp, an implicit grasp policy with an enhanced optimization landscape. This landscape is defined by a NeRF-informed grasp value function. The neural network representing this function is trained on simulated grasp demonstrations. During training, we use an auxiliary loss to guide not only the weight updates of this network but also the slope of the optimization landscape. This loss is computed on the demonstrated grasp trajectory and the gradients of the landscape. It requires second order optimization during training to incorporate valuable information from the trajectory and leads to facilitating the optimization process of the implicit policy. Experiments demonstrate that employing this auxiliary loss improves policies’ performance in simulation as well as their zero-shot transfer to the real-world.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"186 ","pages":"Article 104921"},"PeriodicalIF":4.3,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095621","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}

{"title":"Three-dimensional magnetometer calibration using higher order models embedded into neural networks","authors":"Jon Cowart III ,&nbsp;Jan Petrich","doi":"10.1016/j.robot.2024.104903","DOIUrl":"10.1016/j.robot.2024.104903","url":null,"abstract":"<div><div>This paper proposes a complete solution for magnetometer calibration in three dimensions. Motivated by the classical model containing calibration parameters such as hard and soft iron, we introduce a generalized, multi-variate polynomial model of arbitrary order using higher rank tensor algebra. We highlight the equivalence between a first order polynomial and the classical calibration model. For higher order models, sensor calibration, i.e. the identification of calibration parameters, is carried out using a neural network (NN). The NN is constructed such that the underlying architecture embeds the projection properties of the generalized calibration model, while proper <span><math><msub><mi>L</mi><mn>2</mn></msub></math></span> regularization applied to the calibration parameters minimizes rotational ambiguity of the solution. Experimental validation is carried out using a small MEMS Razor IMU. Results highlight the benefits of the proposed NN approach over traditional calibration methods even for the classical, first order systems. In addition, cross-validation trials reveal the inherent tradeoff between calibration performance and model complexity. The proposed framework is also applicable to accelerometer calibration.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"186 ","pages":"Article 104903"},"PeriodicalIF":4.3,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095625","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":"A design procedure for a novel concept of shape-memory-alloy-actuated finger exoskeleton","authors":"Elio Matteo Curcio ,&nbsp;Giuseppe Carbone","doi":"10.1016/j.robot.2025.104919","DOIUrl":"10.1016/j.robot.2025.104919","url":null,"abstract":"<div><div>This paper provides a thorough investigation demonstrating the feasibility of an innovative finger exoskeleton design concept in a compact, self-contained shape using Shape Memory Alloy (SMA) wires. This novel design integrates SMA wires directly onto the finger body, eliminating the need for bulky wrist actuators. Strategically positioned, the SMA wires serve as both actuators and sensors. The proposed exoskeleton can be used for rehabilitation purposes or for hand augmentation during manufacturing tasks. To properly design the exoskeleton, multiple simulation models have been developed. Kinematic and dynamic analyses were conducted to determine the sufficient motion ranges and forces as a function of SMA wire locations. Both models and experiments confirm the feasibility of integrating SMA wires within the limited space near the fingers. Several experimental tests were conducted to validate the simulation model results. Moreover, thermal-camera measurements confirm that SMA wires can be safely isolated and attached to the finger, preventing skin overheating. All simulation and experimental outcomes indicate the reliability of the proposed design procedure and the engineering feasibility of our novel finger.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"187 ","pages":"Article 104919"},"PeriodicalIF":4.3,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143176576","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":"Formation tracking theory and experiment for leader-following quadrotor swarm under optical motion capture localization system","authors":"Hongtao Dang ,&nbsp;Shijing Guo ,&nbsp;Jianxiang Xi ,&nbsp;Feng Wang ,&nbsp;Cheng Wang ,&nbsp;Yuanshi Zheng","doi":"10.1016/j.robot.2025.104918","DOIUrl":"10.1016/j.robot.2025.104918","url":null,"abstract":"<div><div>The formation tracking approach and the flight experiment for a quadrotor swarm with the leader-following interactive topology are investigated, where it is required that following quadrotors with a specific geometric structure track the state of the leading quadrotor. Firstly, by dividing the interactive topology into the leader-following interactive relationship and the following interactive relationship, the control strategies of the leading and following quadrotors are proposed, respectively. Then, a formation tracking achievability criterion for a quadrotor swarm is proposed by constructing a new Lyapunov function, where two control gains are given in an analytic form. Finally, an algorithm for a quadrotor swarm to realize formation tracking is proposed, where the control strategy is used to determine the outer-loop position control quantity, and a formation tracking flight experiment is performed under the indoor optical motion capture location system to demonstrate the effectiveness of theoretical conclusions.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"186 ","pages":"Article 104918"},"PeriodicalIF":4.3,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095624","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":"Semi-autonomous unmanned aerial manipulator teleoperation for push-and-slide inspection using parallel force/vision control","authors":"Simone D’Angelo ,&nbsp;Mario Selvaggio ,&nbsp;Vincenzo Lippiello ,&nbsp;Fabio Ruggiero","doi":"10.1016/j.robot.2024.104912","DOIUrl":"10.1016/j.robot.2024.104912","url":null,"abstract":"<div><div>Performing inspection and maintenance tasks with aerial robots in complex industrial facilities require high levels of maneuverability and dexterity. As full autonomy still struggles to provide robust solutions due to limited adaptability and high development costs, this study explores the paradigm shift towards shared control teleoperation for tilting unmanned aerial manipulators (UAMs). The research initially focuses on integrating onboard camera measurements and interaction force feedback within a parallel force/vision controller for push-and-slide inspection tasks. The control loop lends itself to the development of a semi-autonomous operation architecture that enables a human operator to easily accomplish the task by means of a simple input device. The paper presents a user study evaluating task completion performance with human-in-the-loop control versus fully autonomous execution. Statistical analysis of 20 user experiences provides insights into the levels of autonomy necessary for effective task completion. Among the analyzed control modalities, statistically significant differences arise when the sliding feature is autonomous, denoting it as the most difficult to manually accomplish. The investigation is conducted within a simulated environment to ensure the safety of sensitive instruments and accommodate users with varying levels of expertise. By proposing shared control architectures, this research addresses the challenges of autonomous UAM operations in hazardous industrial environments, highlighting the benefits of human oversight and control in enhancing task efficiency and safety.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"186 ","pages":"Article 104912"},"PeriodicalIF":4.3,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143095626","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}

{"title":"An expedited BDI agent architecture: Improving the responsiveness of agent-based autonomous systems for handling critical situations","authors":"Leandro Buss Becker ,&nbsp;Iago de Oliveira Silvestre ,&nbsp;Jomi Fred Hübner ,&nbsp;Michael Fisher","doi":"10.1016/j.robot.2025.104917","DOIUrl":"10.1016/j.robot.2025.104917","url":null,"abstract":"<div><div>This work presents improvements on BDI agent architectures towards <em>expedited</em> behaviour. Standard and even real-time BDI architectures, due the characteristics of their deliberative processes, can take considerable time deciding what to do, and this can present a significant delay when a prompt reaction is required in a critical situation. To address this issue without simply adding a pure reaction layer, we introduce a novel, expedited, BDI architecture capable of maintaining effective reasoning while providing adequate and fast reaction to perceptions that occur in a critical situation (E2BA). By ‘adequate’ we mean that a proper action is promptly decided and that the agent temporarily enters into an exceptional operation mode. In this paper we present a concrete implementation for our proposal as a variation of Jason, the leading BDI programming framework. Considering that our main target applications lay in the domain of robotics, we also make our implementation suitable to be used in conjunction with the Robot Operating System (ROS). We evaluate the proposed mechanism through two experiments. The first experiment shows significant (at least 2.7x) reaction-time improvements obtained from using our expedited Jason in comparison with ‘standard’ Jason. Moreover, the results show that expedited Jason reaction-times are not significantly affected as the agent gets busier (it is constant), in contrast to the exponentially increasing response-time of standard Jason. The second experiment addresses the usage of the proposed architecture within a realistic software-in-the-loop (SIL) application scenario: controlling an uncrewed aerial vehicle (UAV) on a fire-fighting mission. This second experiment is not only important for its realistic nature and for reinforcing the reaction-time improvements, but also because it illustrates the importance of changing the agent’s operation mode when performing a failsafe procedure. The permanent overhead introduced by EB2A is very small, not more than 0.5% of the standard BDI reasoning cycle time.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"186 ","pages":"Article 104917"},"PeriodicalIF":4.3,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143096370","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}

{"title":"A dynamic and static combined camera-IMU extrinsic calibration method based on continuous-time trajectory estimation","authors":"Wenlan Ouyang ,&nbsp;Wanbiao Lin ,&nbsp;Lei Sun","doi":"10.1016/j.robot.2025.104916","DOIUrl":"10.1016/j.robot.2025.104916","url":null,"abstract":"<div><div>The accuracy of extrinsic calibration between the camera and IMU is crucial for precise localization and mapping in visual-inertial navigation systems. However, traditional camera-IMU extrinsic calibration methods generally overlook the respective characteristics of the camera and inertial sensor, usually calibrating the entire platform as a whole. Accordingly, while rapid motion can adequately activate the IMU, the image blurring and loss of the visual field caused by excessive motion would affect the accuracy of calibration to some extent. To address this problem, we propose a dynamic and static combined camera-IMU extrinsic calibration method based on continuous-time trajectory estimation. The proposed method introduces a motion capture system as a bridge and skillfully decouples the calibration system into two subsystems, dynamic and static. Therefore, the restrictions between the camera and IMU could be circumvented, with activating IMU sufficiently and avoiding the motion blurring of images, simultaneously. A series of simulations and real-world experiments were conducted to evaluate the effectiveness and accuracy of the proposed calibration method. Moreover, the result was compared with other camera-IMU calibration methods and applied in real-world SLAM to verify the reliability. All the results demonstrate that the proposed method could achieve a more reliable extrinsic with higher precision and consistency.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"186 ","pages":"Article 104916"},"PeriodicalIF":4.3,"publicationDate":"2025-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143135950","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}