Robotics and Autonomous Systems最新文献

{"title":"Cooperative and Asynchronous Transformer-Based Mission Planning for heterogeneous teams of mobile robots","authors":"Milad Farjadnasab,&nbsp;Shahin Sirouspour","doi":"10.1016/j.robot.2025.105131","DOIUrl":"10.1016/j.robot.2025.105131","url":null,"abstract":"<div><div>Cooperative mission planning for heterogeneous teams of mobile robots presents a unique set of challenges, particularly when operating under communication constraints and limited computational resources. To address these challenges, we propose the Cooperative and Asynchronous Transformer-based Mission Planning (CATMiP) framework, which leverages multi-agent reinforcement learning (MARL) to coordinate distributed decision making among agents with diverse sensing, motion, and actuation capabilities, operating under sporadic ad hoc communication. A Class-based Macro-Action Decentralized Partially Observable Markov Decision Process (CMacDec-POMDP) is also formulated to effectively model asynchronous decision-making for heterogeneous teams of agents. The framework utilizes an asynchronous centralized training and distributed execution scheme, enabled by the proposed Asynchronous Multi-Agent Transformer (AMAT) architecture. This design allows a single trained model to generalize to larger environments and accommodate varying team sizes and compositions. We evaluate CATMiP in a 2D grid-world simulation environment and compare its performance against planning-based exploration methods. Results demonstrate CATMiP’s superior efficiency, scalability, and robustness to communication dropouts and input noise, highlighting its potential for real-world heterogeneous mobile robot systems. The code is available at <span><span>https://github.com/mylad13/CATMiP</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"194 ","pages":"Article 105131"},"PeriodicalIF":5.2,"publicationDate":"2025-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144739225","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 novel closed-form solution for automatic calibration of extrinsic parameters based on plane-induced homography","authors":"Beiwei Zhang,&nbsp;Zhiwang Zhang,&nbsp;Xueru Gu","doi":"10.1016/j.robot.2025.105144","DOIUrl":"10.1016/j.robot.2025.105144","url":null,"abstract":"<div><div>The extrinsic calibration of the camera is an important task for deploying an autonomous device. This paper proposes a novel method for the estimation of the six degrees of freedom extrinsic parameters in a camera-robot system, in which the plane-induced Homography between two camera images is employed. The characteristics of the eigenvalues and eigenvectors in the Homography are firstly investigated and identified supposing the robot performs a known translation motion. Then a closed-form solution for rotation matrix can be achieved from the eigenvectors and so is the translation vector in the extrinsic parameters. Three simulation experiments are performed to validate the correctness, accuracy and robustness. Besides, one real data experiment is implemented on our robot platform. For comparison, two recently reported algorithms are carried out under similar settings. The experimental results of means and standard deviations for those parameters demonstrate that the proposed solution can provide satisfactory performance.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"194 ","pages":"Article 105144"},"PeriodicalIF":5.2,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144748620","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":"Data efficient Deep Reinforcement Learning for robust inertial-based UAV localization","authors":"Dimitrios Tsiakmakis ,&nbsp;Nikolaos Passalis ,&nbsp;Anastasios Tefas","doi":"10.1016/j.robot.2025.105139","DOIUrl":"10.1016/j.robot.2025.105139","url":null,"abstract":"<div><div>Precise localization is a critical task for many UAV-based applications. Inertial Measurement Units (IMUs), which measure acceleration and angular velocity, are commonly used for UAV localization due to their low cost and small size. However, IMU-based localization is prone to accumulating errors over time, which can significantly impact the accuracy of the localization. To address this issue, we propose a data efficient Deep Reinforcement Learning (DRL) method that enables learning how to correct localization errors from IMUs. Our approach utilizes a novel data augmentation method, along with an appropriate “hint” loss that can provide additional supervision during the training process. As a result, the proposed method requires a very small number of real-world examples and can be implemented using widely available low cost RGB sensors, ensuring that it can be readily applied in a wide range of different applications. We demonstrate the effectiveness of the proposed method in both simulation and real-world UAV experiments. In comparison to traditional supervised and DRL approaches, the proposed approach allows for achieving more precise localization with fewer real-world examples, making it a practical tool for adapting DL-based localization models for UAV applications.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"194 ","pages":"Article 105139"},"PeriodicalIF":5.2,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144724224","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":"Unconventional grasping in multi-limb humanoid robot using fuzzy logic intelligence","authors":"Indramani,&nbsp;Arun Dayal Udai,&nbsp;Sanjoy K. Ghoshal","doi":"10.1016/j.robot.2025.105129","DOIUrl":"10.1016/j.robot.2025.105129","url":null,"abstract":"<div><div>This research article outlines grasping using the whole body of a human and explores the similar grasping potential of a humanoid robot. The unconventional grasping for a humanoid robot is addressed, in which the robot can perform autonomous grasping by the potential body limbs based on the object’s location and size. A comprehensive grasping possibility with human body limbs is studied using a stick diagram. These limb-assisted grasps are reported to utilize the theory of enumerative combinatorics, and an upper-body humanoid robot is used to implement some feasible intelligent, unconventional grasping. Single-Input Multi-Output and Multi-Input Multi-Output fuzzy logic control algorithms are developed to acquire the intelligent grasp objectives. These algorithms assign grasp weightage based on the size and position of the object to instruct the individual robotic arms to be involved in the grasping. Further, a Humanoid Robot Grasp Control algorithm is developed for head and arm motion control that assists in grasp execution. Finally, unconventional grasps are implemented using an upper-body dual-arm humanoid robot consisting of a 7 DoF robot with left and right arms, torso, and head in the virtual environment. The proposed strategy offers a benefit in intelligent grasping and improves the adaptability of a humanoid robot’s limbs for handling a variety of primitive-shaped objects.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"194 ","pages":"Article 105129"},"PeriodicalIF":5.2,"publicationDate":"2025-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144722891","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":"Resource-constrained dynamic planning and model-free control in turbulent urban environments","authors":"Thomas Nakken Larsen ,&nbsp;Mandar Tabib ,&nbsp;Adil Rasheed","doi":"10.1016/j.robot.2025.105130","DOIUrl":"10.1016/j.robot.2025.105130","url":null,"abstract":"<div><div>This work proposes a descriptive digital twin of an urban region with a sufficiently small footprint to run online in resource-constrained quadrotor systems. The digital twin describes the building geometries and turbulent kinetic energy (TKE) within an urban region, enabled via our proposed surrogate model (SM) for TKE reconstruction. This forms the basis of a simulation environment for autonomous path following and collision avoidance using Deep Reinforcement Learning (DRL). A Voronoi-based turbulence-weighted graph (TWG) is developed for safe path planning and is capable of reacting to dynamic changes in wind direction and, consequently, the turbulence field. Lastly, the environment simulates oncoming traffic of dynamic unknown obstacles for a vision-enabled DRL agent to evade. Several DRL agents with different observation and action spaces are trained and evaluated.</div><div>The SM enables rapid reconstruction of the turbulence with an accuracy comparable to full-order methods. The TWG plans safe paths that reduce the worst-case TKE exposure by 44% at the cost of increasing the average path length by 33% compared to a shortest-distance approach. The DRL agents successfully solve the navigation problem with a 100% success rate in the static obstacle scenario, where the minimum clearance between buildings is <span><math><mrow><mn>1</mn><mo>.</mo><mn>0</mn><mi>m</mi></mrow></math></span>. In scenarios with dynamic obstacles, the agent achieves a 79% success rate. Suggestions for further performance and safety improvements in the TWG planner and DRL agents are presented.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"194 ","pages":"Article 105130"},"PeriodicalIF":5.2,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144724292","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":"Very large-scale multi-robot task allocation in challenging environments via robot redistribution","authors":"Seabin Lee,&nbsp;Joonyeol Sim,&nbsp;Changjoo Nam","doi":"10.1016/j.robot.2025.105126","DOIUrl":"10.1016/j.robot.2025.105126","url":null,"abstract":"<div><div>We consider the Multi-Robot Task Allocation (MRTA) problem that aims to optimize an assignment of multiple robots to multiple tasks in challenging environments which are with densely populated obstacles and narrow passages. In such environments, conventional methods optimizing the sum-of-cost are often ineffective because the conflicts between robots incur additional costs (e.g., collision avoidance, waiting). Also, an allocation that does not incorporate the actual robot paths could cause deadlocks, which significantly degrade the collective performance of the robots.</div><div>We propose a scalable MRTA method that considers the paths of the robots to avoid collisions and deadlocks which result in a fast completion of all tasks (i.e., minimizing the <em>makespan</em>). To incorporate robot paths into task allocation, the proposed method constructs a roadmap using a Generalized Voronoi Diagram. The method partitions the roadmap into several components to know how to redistribute robots to achieve all tasks with less conflicts between the robots. In the redistribution process, robots are transferred to their final destinations according to a push-pop mechanism with the first-in first-out principle. From the extensive experiments, we show that our method can handle instances with hundreds of robots in dense clutter while competitors are unable to compute a solution within a time limit.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"194 ","pages":"Article 105126"},"PeriodicalIF":5.2,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144722901","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":"AEROBULL: A Center-of-Mass displacing aerial vehicle enabling efficient high-force interaction","authors":"Tong Hui ,&nbsp;Stefan Rucareanu ,&nbsp;Esteban Zamora ,&nbsp;Simone D’Angelo ,&nbsp;Haotian Liu ,&nbsp;Matteo Fumagalli","doi":"10.1016/j.robot.2025.105127","DOIUrl":"10.1016/j.robot.2025.105127","url":null,"abstract":"<div><div>Aerial manipulators are increasingly used in contact-based industrial applications, where tasks like drilling and pushing require platforms to exert significant forces in multiple directions. To enhance force generation capabilities, various approaches, such as thrust vectoring and perching, have been explored. In this article, we introduce a novel approach by investigating the impact of varied CoM (Center of Mass) locations on an aerial manipulation system’s force exertion. Our proposed platform-AEROBULL features a design with a dynamically displacing CoM, enabling a smooth transition between free flight and high-force interactions supported by tilting back rotors. We provide a detailed study of the aerial platform’s overall system design, hardware integration of the developed physical prototype, and software architecture of the proposed control algorithm. Physical experiments were conducted to validate the control design and explore the force generation capability of the designed platform via a pushing task. With a total mass of <span><math><mrow><mn>3</mn><mo>.</mo><mn>12</mn><mspace></mspace><mstyle><mi>k</mi><mi>g</mi></mstyle></mrow></math></span>, the aerial vehicle exerted a maximum pushing force of above <span><math><mrow><mn>28</mn><mspace></mspace><mstyle><mi>N</mi></mstyle></mrow></math></span> being almost equal to its gravity force. Furthermore, the experiments illustrated the benefits of having displaced CoM by benchmarking with a fixed CoM configuration. Additionally, we introduce a quantitative factor to compare the force exertion capabilities of our system with existing platforms, highlighting the advantages of our approach.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"194 ","pages":"Article 105127"},"PeriodicalIF":4.3,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144712998","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 barrier Lyapunov-based fast adaptive robust control system for 6-DOF autonomous submersible vehicles","authors":"Hossein Ahmadian ,&nbsp;Mohammad Mehdi Arefi ,&nbsp;Alireza Khayatian ,&nbsp;Allahyar Montazeri","doi":"10.1016/j.robot.2025.105115","DOIUrl":"10.1016/j.robot.2025.105115","url":null,"abstract":"<div><div>This paper proposes a novel fast adaptive back-stepping robust controller, based on the<em>barrier Lyapunov function</em> (<span><math><mi>BLF</mi></math></span>), to address the position and velocity constraints typically imposed in the design of <em>Euler–Lagrange systems</em>. The aim is to improve upon various aspects of conventional <span><math><msub><mrow><mi>L</mi></mrow><mrow><mn>1</mn></mrow></msub></math></span> adaptive control and Model Reference Adaptive Control (<span><math><mi>MRAC</mi></math></span>). The proposed controller reduces complexity by eliminating the low-pass filter from the design process in <span><math><msub><mrow><mi>L</mi></mrow><mrow><mn>1</mn></mrow></msub></math></span> adaptive control, resulting in faster convergence and enhanced robustness against <em>nonlinear uncertainties, external disturbances, and actuator dynamics</em>, which are crucial in real-world applications. The performance of the proposed scheme is evaluated on two different Euler–Lagrange systems: a 6-degree-of-freedom (6-<span><math><mi>DOF</mi></math></span>) remotely operated vehicle (<span><math><mi>ROV</mi></math></span>) and a single-link robot manipulator. Key performance indicators such as settling time, overshoot percentage, control effort, and trajectory tracking error are used for assessment. The results confirm that the proposed controller outperforms both <span><math><msub><mrow><mi>L</mi></mrow><mrow><mn>1</mn></mrow></msub></math></span> adaptive control and <span><math><mi>MRAC</mi></math></span> in terms of tracking accuracy and state estimation errors for both position and velocity outputs. Additionally, the proposed method demonstrates superior performance in handling actuator dynamics, mitigating matched nonlinear time-varying disturbances, and achieving precise trajectory tracking, even in the presence of input gain uncertainties. These improvements establish the proposed controller as a more robust and efficient alternative to traditional adaptive control methods.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"194 ","pages":"Article 105115"},"PeriodicalIF":5.2,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144722902","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":"Enhanced Hierarchical Fuzzy Formation Control with fuzzy collision avoidance behavior for multiple Mecanum wheeled Mobile Robots","authors":"Hsiu-Ming Wu ,&nbsp;Muhammad Qomaruz Zaman","doi":"10.1016/j.robot.2025.105124","DOIUrl":"10.1016/j.robot.2025.105124","url":null,"abstract":"<div><div>Integrating collision avoidance mechanisms into formation control represents a critical aspect for enabling multi-mobile robotic coordination from arbitrary initial configurations. Additionally, the reliance on precise system models for controller design and leader-centralized control architecture limits the flexibility to dynamically reconfigure the formation structure, which is often crucial in real-world applications. Consequently, exploration of modularly interpretable and model-free control strategies emerge as a compelling research direction to address contemporary robotic coordination challenges. This study proposes a Hierarchical Fuzzy Formation Control (HFFC) approach for multiple Mecanum-wheeled Mobile Robots (MMRs) to achieve simultaneous formation tracking, collision avoidance, and orientation alignment. The HFFC leverages a modular hierarchical fuzzy inference system, combining leader–follower and behavior-based strategies. Fuzzified sliding surfaces enhance the tracking performance by minimizing oscillation and chattering effects during formation convergence. Collision avoidance is improved by incorporating inter-MMR approaching rate, enabling proactive anticipation and more responsive maneuvers. A realistic Takagi–Sugeno model, replicating real-world MMR behavior with practical actuator voltage inputs, is developed for evaluation. Simulations demonstrate that five MMRs achieve the desired formation geometry within 1.9 s with 0.048 m accuracy while maintaining a minimum inter-robot distance of 10.77 cm to prevent collisions. Moreover, compared to existing approaches, the proposed control scheme possesses better performance.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"194 ","pages":"Article 105124"},"PeriodicalIF":4.3,"publicationDate":"2025-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144679720","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":"Path and footfall planning for N-legged and climbing robots — A model predictive control approach","authors":"Carlos Prados,&nbsp;Miguel Hernando,&nbsp;Ernesto Gambao","doi":"10.1016/j.robot.2025.105119","DOIUrl":"10.1016/j.robot.2025.105119","url":null,"abstract":"<div><div>In this paper, we present a general control framework for N-legged and variably-configured robots, designed to coordinate leg movements for climbing tasks without relying on Central Pattern Generators (CPGs). Model-based path and footfall planners are introduced to minimize actuator effort, minimize robot detachment risk, improve payload distribution between legs, and maximize the traveled distance during the swing phase. To achieve this, we address the force distribution problem (FDP) by selecting configurations where the robot is most comfortable in terms of kinematics, effort, and safety. A gait controller is presented as a nonperiodic, nonsymmetric, and nonregular bioinspired method that selects the most convenient leg to move by ensuring comfort, safety, and robot capabilities. The system has been tested in simulation with different robot configurations (varying number of legs and arrangements) and with the physical robot ROMERIN in its quadruped version.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"194 ","pages":"Article 105119"},"PeriodicalIF":4.3,"publicationDate":"2025-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144679718","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}