Journal of Field Robotics最新文献

{"title":"Cover Image, Volume 42, Number 5, August 2025","authors":"Hongchuan Zhang,&nbsp;Junkai Ren,&nbsp;Junhao Xiao,&nbsp;Hainan Pan,&nbsp;Huimin Lu,&nbsp;Xin Xu","doi":"10.1002/rob.70036","DOIUrl":"https://doi.org/10.1002/rob.70036","url":null,"abstract":"<p>The cover image is based on the article <i>FTR-bench: Benchmarking deep reinforcement learning for flipper-track robot control</i> by Huimin Lu et al., 10.1002/rob.22528.\u0000\u0000 <figure>\u0000 <div><picture>\u0000 <source></source></picture><p></p>\u0000 </div>\u0000 </figure></p>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 5","pages":""},"PeriodicalIF":4.2,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/rob.70036","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144680976","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":"FTR-Bench: Benchmarking Deep Reinforcement Learning for Flipper-Track Robot Control","authors":"Hongchuan Zhang,&nbsp;Junkai Ren,&nbsp;Junhao Xiao,&nbsp;Hainan Pan,&nbsp;Huimin Lu,&nbsp;Xin Xu","doi":"10.1002/rob.22528","DOIUrl":"https://doi.org/10.1002/rob.22528","url":null,"abstract":"<div>\u0000 \u0000 <p>Tracked robots equipped with flippers and sensors are extensively employed in outdoor search and rescue scenarios. However, achieving precise motion control on complex terrains remains a significant challenge, often necessitating expert teleoperation. This stems from the high degree of robot joint freedom and the need for precise flipper coordination based on terrain roughness. To address this problem, we propose <span>F</span>lipper-\u0000<span>T</span>rack \u0000<span>R</span>obot \u0000<span>Bench</span> mark (<b>FTR-Bench</b>), a simulator featuring flipper-track robots tasked with crossing various obstacles using reinforcement learning (RL) algorithms. The primary objective is to enable autonomous locomotion in environments that are too remote or hazardous for humans, such as disaster zones or planetary surfaces. Built on Isaac Lab, FTR-Bench achieves efficient RL training at over 4000 FPS on an RTX 3070 GPU. Additionally, it integrates RL algorithms with OpenAI Gym interface specifications, enabling fast secondary development and verification. On this basis, FTR-Bench provides a series of standardized RL-based benchmarking experiments baselines for obstacle-crossing tasks, providing a solid foundation for subsequent algorithm design and performance comparison. Experimental results empirically indicate that SAC algorithms performs relatively well in single and mixed terrain traversal, but most algorithms struggle with multi-terrain traversal skills, which calls the RL community for more substantial development. Our project is open-source at https://github.com/nubot-nudt/FTR-Benchmark.</p></div>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 5","pages":"2375-2389"},"PeriodicalIF":4.2,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681151","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":"MoWe: Motion Observation for Wind Estimation of Sailing Robots","authors":"Qinbo Sun,&nbsp;Weimin Qi,&nbsp;Huihuan Qian","doi":"10.1002/rob.22512","DOIUrl":"https://doi.org/10.1002/rob.22512","url":null,"abstract":"<div>\u0000 \u0000 <p>Toward sustained mobility in complex marine environment, there is an urgent need for sailing robots to operate robustly when wind measurements are unavailable (e.g., wind sensors are damaged). This study proposes an effective motion observation-based wind estimation (MoWe) scheme, which enables the robotic sailboat to consistently acquire wind information from its own maneuvers. MoWe incorporates motion analysis (MA) and data-driven (DD) methods. In the MA method, dead zone constraints of the robotic sailboat are identified as crucial references in deriving wind direction. For the DD approach, the sailing robot as shown in Figure 1 is employed to collect a data set, which serves as the basis for regressing an estimator. We conducted extensive validation tests in both simulation and experiments. Results indicate favorable performance for both methods in simulated scenarios. Notably, the DD method exhibited higher estimation accuracy in all experiments. The mean absolute error (MAE) of estimated wind direction was 4.25°, with the range of confidence interval spanning from 25.13° to 33.56°, demonstrating the robustness of the DD method. Furthermore, the estimation of wind direction has been successfully applied in straight-line sailing tests, and the wind magnitude has been estimated. The MAE of wind magnitude estimation was 1.13m/s.</p></div>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 5","pages":"2355-2374"},"PeriodicalIF":4.2,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681152","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":"Real-time Nonlinear Model Predictive Control of a Robotic Arm Using Spatial Operator Algebra Theory","authors":"Tuğçe Yaren,&nbsp;Selçuk Kizir","doi":"10.1002/rob.22514","DOIUrl":"https://doi.org/10.1002/rob.22514","url":null,"abstract":"<p>Nonlinear model predictive control (NMPC) has inherent challenges, such as high computational burden, nonconvex optimization, and the necessity of powerful and fast processors with large memory for real-time robotics. In this study, a new NMPC strategy is proposed using Spatial Operator Algebra (SOA) theory to address these challenges, and experimental results are presented for the five degrees of freedom robot manipulator. The proposed scheme is based on an NMPC controller using the SOA-based dynamic model to provide good tracking performance and ensure the satisfaction of constraints. Two novel control schemes, SOA–NMPC and SOA–NMPC proportional-derivative (PD), are introduced for a comprehensive analysis of the proposed innovative approach. The validity of the proposed scheme is experimentally tested through robustness analysis conducted across various tasks, including the addition of weight and exposure to internal/external disturbances. The effectiveness of the proposed approach is demonstrated through benchmarking against NE-NMPC using the Newton–Euler (NE) algorithm, classical MPC, MPC-PD, and PID techniques. The comparative results show that the SOA–NMPC controller provides effective performance and ensures constraints for the entire trajectory of the manipulator, even under varying conditions.</p>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 5","pages":"2337-2354"},"PeriodicalIF":4.2,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/rob.22514","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681571","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":"Improving Localization Accuracy of Offline Navigation Algorithms for Intelligent Pipeline Inspection Gauges and In-Line Inspection Robotic Systems","authors":"Kwanghyun Yoo,&nbsp;Dae-Kwang Kim,&nbsp;Jae-Jun Kim,&nbsp;Seung-Ung Yang,&nbsp;Hui-Ryoung Yoo,&nbsp;HongSeok Song,&nbsp;Han-You Jeong,&nbsp;Hoa-Hung Nguyen,&nbsp;Jin Woo Song,&nbsp;Dong-Kyu Kim","doi":"10.1002/rob.22501","DOIUrl":"https://doi.org/10.1002/rob.22501","url":null,"abstract":"<div>\u0000 \u0000 <p>Integrity management of pipeline networks is crucial for preemptive maintenance and preventing accidents. In this study, various methods for improving the localization accuracy of the offline navigation algorithm for an intelligent pipeline inspection gauge (PIG) and in-line inspection (ILI) robotic system are proposed. The digital mapping algorithm utilizes the extended Kalman filter (EKF) with a Rauch–Tung–Striebel (RTS) smoother. Two ILI tools are introduced—magnetic flux leakage (MFL) PIG, which is a typical intelligent PIG for detecting corrosion defects in pipelines, and a new low-friction geometry robot (LFGR) for inspecting the mechanical defects in low-pressure, low-flow pipelines. The MFL PIG has only three odometers to measure the speed of the PIG along the moving direction. Hence, a compensation method for the measured speed was developed and utilized. In addition, an optimization procedure for the parameters of sensor uncertainty modeling was proposed and validated. These methods increased the localization accuracy of the digital mapping algorithm of the MFL PIG. Specifically, the root mean squared value of the two-dimensional distance error decreased by 47.73%. The proposed methods were applied to the LFGR equipped with four odometers and a high-accuracy inertial measurement unit. Moreover, additional sensors and a new algorithm for attitude angle correction of the robot were utilized. The proposed methods were successfully validated using field ILI results. The methods enhance the effectiveness of the integrity management of pipeline systems, thus contributing toward their safe and reliable operation.</p>\u0000 </div>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 5","pages":"2218-2233"},"PeriodicalIF":4.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681630","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":"Cleaning Robots: A Review of Sensor Technologies and Intelligent Control Strategies for Cleaning","authors":"Rajesh Kannan Megalingam,&nbsp;Shree Rajesh Raagul Vadivel,&nbsp;Sai Smaran Kotaprolu,&nbsp;Bagathi Nithul,&nbsp;Devisetty Vijay Kumar,&nbsp;Gaurav Rudravaram","doi":"10.1002/rob.22515","DOIUrl":"https://doi.org/10.1002/rob.22515","url":null,"abstract":"<div>\u0000 \u0000 <p>Cleaning robots have revolutionized the way spaces and surfaces are cleaned, accessing areas that are often difficult for humans to reach. A significant segment of the cleaning robot industry is dedicated to home cleaning robots, which have undergone numerous enhancements since their initial introduction. Through extensive research and development efforts, the application scope of cleaning robots has expanded beyond floors to encompass walls, staircases, pools, tanks, ventilation ducts, windows, and other surfaces. These robots have seen improvements in cleaning effectiveness through the integration of new designs, technologies, and control methods. This review paper presents a comprehensive analysis of cutting-edge cleaning robots recently developed, exploring various sensor technologies and intelligent control strategies employed to enhance cleaning efficiency. Additionally, the paper discusses the challenges associated with deploying these robots in real-world scenarios.</p>\u0000 </div>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 5","pages":"2234-2259"},"PeriodicalIF":4.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681231","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 Spiral Coverage Path Planning Algorithm for Nonomnidirectional Robots","authors":"Taogang Hou,&nbsp;Jiaxin Li,&nbsp;Xuan Pei,&nbsp;Hao Wang,&nbsp;Tianhui Liu","doi":"10.1002/rob.22516","DOIUrl":"https://doi.org/10.1002/rob.22516","url":null,"abstract":"<div>\u0000 \u0000 <p>The limited steering capabilities of nonomnidirectional robots introduce significant complexity into complete coverage tasks, often leading to increased path overlap or incomplete coverage of certain areas. Although recent research has made progress in optimizing coverage path planning, redundant coverage or omissions are still prone to occur in the target area to be covered. To address these persistent challenges, we propose a novel spiral coverage method. This approach not only conforms to the kinematic constraints of nonomnidirectional robots but also enhances coverage efficiency by dividing the target area into center and boundary regions and devising tailored coverage strategies for each. This method effectively reduces path redundancy and improves overall area coverage. Furthermore, we introduce a comprehensive metric that combines total path length and area coverage ratio to evaluate coverage efficiency, overcoming the limitations and computational complexity associated with existing metrics. For scenarios where maximizing the area coverage ratio is critical, we have developed a high-coverage-rate turning strategy that ensures 100% coverage. Through simulation tests in six representative areas and actual experiments on airport runways, our method shows an improvement of 0.238%–14.538% in coverage efficiency compared with parallel coverage method and 60.548%–76.339% compared with deep reinforcement learning-based method. Additionally, implementing high-coverage-rate turning strategies improves the area coverage ratio by 2.021%–6.732%. In field experiments, our method reduces execution time by 1.61% compared with parallel coverage method. These results show that our method has a significant effect in improving coverage efficiency and achieving complete coverage goals.</p></div>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 5","pages":"2260-2279"},"PeriodicalIF":4.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681232","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 Universal Visual Detection Method for Camellia oleifera Fruit Picking Robot","authors":"Jinpeng Wang,&nbsp;Fei Yuan,&nbsp;Jialiang Zhou,&nbsp;Meng He,&nbsp;Qianguang Zhen,&nbsp;Chenzhe Fang,&nbsp;Sunan Chen,&nbsp;Hongping Zhou","doi":"10.1002/rob.22518","DOIUrl":"https://doi.org/10.1002/rob.22518","url":null,"abstract":"<div>\u0000 \u0000 <p>In recent years, the application of robots in the field of fruit picking has steadily increased. Mechanized methods for harvesting oil tea fruits include comb picking, vibratory picking, and gripping picking, among others. Traditional reliance on a single-picking method is limited by variability in fruit size, shading, and environmental conditions. To develop a universal vision system suitable for picking robots capable of multiple picking methods and achieve intelligent harvesting of oil tea fruits. This paper proposes an enhanced You Only Look Once v7 (YOLOv7)-based oil tea fruits recognition method specifically designed for subsequent clamp or comb picking. The network's feature extraction capability is enhanced by incorporating an attention mechanism, an optimized small target detection layer, and an improved training loss function, thereby improving its detection of occluded and small target fruits. An innovative Automatic Assignment (AA) method clusters and subclusters the detected oil tea fruits, providing crucial fruit distribution data to optimize the robot's picking strategy. Additionally, for vibration harvesting, this paper introduces a vibration point detection method utilizing the Pyramid Scene Parsing Network (PSPNet) semantic segmentation network combined with connectivity domain analysis to identify vibration points on the trunks and branches of oil tea trees. Experimental results demonstrate that the generalized visual detection method proposed in this study surpasses existing models in identifying oil tea fruit trees, with the enhanced YOLOv7 model achieving mean average precision, recall, and accuracy of 91.7%, 94.0%, and 94.9%, respectively. The AA method achieves clustering and subclustering of oil tea fruits with a processing delay of under 5 ms. For vibration harvesting, PSPNet achieves branch segmentation precision, recall, and intersection ratio of 97.3%, 96.5%, and 94.5%, respectively. The proposed branch vibration point detection method attains a detection accuracy of 93%, effectively pinpointing vibration points on the trunks and branches of oil tea trees. Overall, the proposed visual method can be implemented in robots using various picking techniques to enable automated harvesting.</p>\u0000 </div>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 5","pages":"2280-2296"},"PeriodicalIF":4.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681627","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":"Toward Universal Embodied Planning in Scalable Heterogeneous Field Robots Collaboration and Control","authors":"Hanwen Wan,&nbsp;Yuhan Zhang,&nbsp;Junjie Wang,&nbsp;Donghao Wu,&nbsp;Mengkang Li,&nbsp;Xilun Chen,&nbsp;Yixuan Deng,&nbsp;Yuxuan Huang,&nbsp;Zhenglong Sun,&nbsp;Lin Zhang,&nbsp;Xiaoqiang Ji","doi":"10.1002/rob.22522","DOIUrl":"https://doi.org/10.1002/rob.22522","url":null,"abstract":"<div>\u0000 \u0000 <p>Multi-robot systems offer substantial enhancements in efficiency, scalability, robustness, and flexibility for executing complex tasks through collaborative efforts. However, existing methodologies are constrained by their lack of generalizability, the need for extensive modeling, and most importantly, limitations in their applicability in complex scenarios. This paper presents a novel approach to multi-robot task planning and coordination, introducing a comprehensive pipeline encompassing data generation, supervised fine-tuning, and rigorous error analysis using the Multi-Robot collaboration Error Diagnostic (MRED) metrics. Bridging the gap between natural language commands and physical groundings in robot collaboration tasks, we present <i>MultiPlan</i>: the first data set specifically designed for LLM fine-tuning. The MultiPlan data set encompasses 100 distinct indoor and outdoor scenarios, ranging from office to garden. Experiments underscore the efficacy of the proposed methodology, including comparative analyses against state-of-the-art LLMs and generalization studies on previously unseen tasks. Results reveal that the fine-tuned model achieves a 24.8% relative improvement over the GPT-4 model in addressing complex multi-robot planning scenarios. We also conducted field evaluations in both office and urban settings to demonstrate the deployment performance of the proposed method. These results demonstrate the model's superior capabilities in task decomposition, error management, and adaptation to novel contexts.</p></div>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 5","pages":"2318-2336"},"PeriodicalIF":4.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681629","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":"Multistage Synchronous Telescopic Manipulator With End-Effector–Biased Rotating-Pulling Mode for Damage-Free Robotic Picking","authors":"Xiaoqiang Du,&nbsp;Zhichao Meng,&nbsp;Yawei Wang,&nbsp;Yuechan Li,&nbsp;Zenghong Ma,&nbsp;Leiying He,&nbsp;Wenwu Lu,&nbsp;Jianneng Chen,&nbsp;Chuanyu Wu,&nbsp;Manoj Karkee","doi":"10.1002/rob.22521","DOIUrl":"https://doi.org/10.1002/rob.22521","url":null,"abstract":"<div>\u0000 \u0000 <p>Fruit picking is one of the most time-consuming and labor-intensive stages of fruit production, characterized by high labor demands and significant labor costs. Traditional fruit-picking robotic manipulators typically adopt configurations similar to general-purpose industrial robots, following a predefined path and employing a direct-pulling mode to detach the fruit. However, due to the constraints of the orchard environment and the varying conditions of the fruit, manipulators should be designed to accommodate the specific horticultural characteristics of the trees to improve picking efficiency. Additionally, the picking process should be optimized based on the biological characteristics of the fruit to ensure quality. In this study, a five-degree-of-freedom manipulator based on a multistage synchronous telescopic mechanism is proposed for fruit picking. Workspace analysis indicates that the manipulator can cover more than 80% of the fruit distribution on the trees. To ensure motion accuracy, a FreeRTOS-based motion control system is developed for the manipulator. To evaluate picking efficiency and quality, fruit-picking experiments are conducted in an apple orchard. A rope-driven, three-finger end-effector is mounted in a biased position at the end of the manipulator, complemented by an RGB-D camera for fruit detection and a ROS-based control system for robotic operation. The performance of two picking modes (direct-pulling and biased rotating-pulling) are compared in these experiments. The results demonstrate that the biased rotating-pulling mode yields a higher picking success rate and a lower stem damage rate compared with the direct-pulling mode. Specifically, the damage-free success rate for the biased rotating-pulling mode is 80%, with a 9.18% reduction in stem damage compared with the direct-pulling mode. Furthermore, the average picking cycle time is approximately 14.5 s. In conclusion, the manipulator and its motion control system successfully achieve efficient, nondestructive fruit picking with a high success rate, offering valuable insights for the development of fully automated fruit-picking robots in the future.</p>\u0000 </div>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 5","pages":"2297-2317"},"PeriodicalIF":4.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681628","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}