Article
Automation & Control Systems
Guoxiang Zhao, Minghui Zhu
Summary: This paper investigates the problem of motion planning for multiple unicycle robots to reach their respective goal regions safely and with minimal traveling times. The authors propose a distributed algorithm that combines decoupled optimal feedback planning and distributed conflict resolution. The algorithm guarantees collision avoidance and finite-time arrival at the goal regions. Furthermore, the computational complexity of the algorithm is independent of the number of robots. Simulations are conducted to verify the scalability and near-optimality of the algorithm.
Article
Robotics
Virginia Ruiz Garate, Soheil Gholami, Arash Ajoudani
Summary: This article introduces an online scalable tele-impedance framework for individual and collaborative control of multiple robotic platforms. The framework allows easy addition or removal of robots through the virtual hand concept. Experimental results demonstrate the framework's adaptability to various robotic platforms, the number of robots involved, and task requirements.
IEEE TRANSACTIONS ON ROBOTICS
(2021)
Article
Robotics
Mrinal Verghese, Nikhil Das, Yuheng Zhi, Michael Yip
Summary: Real-time motion planning for robots in complex high-dimensional environments is still an open problem. In this study, we propose a novel approach using K-Means clustering in the Forward Kinematics space to accelerate collision checking. By training individual configuration space models, we obtain compact yet highly accurate models that can be queried rapidly in more complex environments. Experimental results demonstrate that our method, called Decomposed Fast Perceptron (D-Fastron), achieves significantly faster collision checks and motion planning compared to state-of-the-art geometric collision checkers.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Computer Science, Artificial Intelligence
Zheyuan Wang, Chen Liu, Matthew Gombolay
Summary: In this paper, a novel heterogeneous graph attention network model called ScheduleNet is proposed to learn scheduling policies that overcome the limitations of conventional methods. By introducing robot- and proximity-specific nodes into the simple temporal network, a nonparametric heterogeneous graph structure is obtained. The model is shown to be end-to-end trainable on small-scale problems and generalizes to large, unseen problems, outperforming existing state-of-the-art methods in various testing scenarios involving both homogeneous and heterogeneous robot teams.
Article
Computer Science, Information Systems
Bayadir A. Issa, Abdulmuttalib Turky Rashid
Summary: This paper proposes a new approach called the Neighbor-Leader Algorithm for controlling robot formation. The algorithm estimates the positions and orientations of the robots using a localization procedure and controls the formation by rearranging and changing the direction of the neighbor robots. The simulation results demonstrate the effectiveness of the algorithm.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Automation & Control Systems
Leonore Winterer, Sebastian Junges, Ralf Wimmer, Nils Jansen, Ufuk Topcu, Joost-Pieter Katoen, Bernd Becker
Summary: This research focuses on synthesis problems with constraints in POMDPs and proposes an abstraction refinement framework to transform a POMDP model into a probabilistic two-player game, allowing for determination of optimal strategies through efficient verification and synthesis tools. This method advances the state of the art in solving planning problems in partially observable environments with safety guarantees.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2021)
Article
Engineering, Civil
Chengchao Bai, Peng Yan, Wei Pan, Jifeng Guo
Summary: This study proposes a multi-robot adaptive formation control framework based on deep reinforcement learning, comprising an execution layer and a decision-making layer, enabling robots to independently adjust their formation to pass through obstacle areas and apply in various scenarios.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Robotics
Lei Yan, Theodoros Stouraitis, Sethu Vijayakumar
Summary: Multi-robot teams can achieve more complex tasks, but face challenges such as limited communication and uncertain system parameters. A Decentralized Ability-Aware Adaptive Control method is proposed to address these challenges, enabling decentralized coordination and load distribution among robots in collaborative manipulation tasks.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Article
Automation & Control Systems
Xiaohua Ge, Qing-Long Han, Jun Wang, Xian-Ming Zhang
Summary: This paper investigates the problem of distributed formation tracking control and obstacle avoidance of multi-vehicle systems in complex obstacle-laden environments. A neural network is used to model the unknown nonlinearity of vehicle dynamics and repulsive potentials are employed for obstacle avoidance. The proposed approach achieves effective formation tracking control for nonlinear and uncertain multi-vehicle systems without requiring global information.
IEEE-CAA JOURNAL OF AUTOMATICA SINICA
(2022)
Article
Automation & Control Systems
Shiyu Zhang, Federico Pecora
Summary: This paper proposes a method for multi-robot motion coordination, which separates the problem into a global coordination stage and a local trajectory replanning stage. The method achieves online reactivity and scalability through coordination in a reduced space and efficient trajectory replanning. Experiments show that the method has low computational overhead and scales quadratically with the number of robots.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
Article
Robotics
Wil Thomason, Hadas Kress-Gazit
Summary: This article proposes a method for automatically constructing and improving robot action abstractions through observations of the robot's actions. The method, called automatic abstraction repair, uses constrained polynomial zonotopes (CPZs) to model both symbolic and geometric states. The repair process performs an optimizing search to improve the grounding of the abstraction to the behavior of a physical robot.
IEEE TRANSACTIONS ON ROBOTICS
(2023)
Article
Chemistry, Multidisciplinary
Di Liang, Zhongyi Liu, Ran Bhamra
Summary: In this paper, an improved artificial potential field with a dynamic virtual target point and ant colony optimization are combined for multi-robot cooperative formation and global path planning. By adjusting the repulsive field function and introducing a dynamic virtual target point, the local minima and target unreachability problems of the traditional method are solved, resulting in improved convergence speed and global optimization accuracy of the algorithm.
APPLIED SCIENCES-BASEL
(2022)
Article
Engineering, Multidisciplinary
Jiabao Wen, Jiachen Yang, Yang Li, Jingyi He, Zhengjian Li, Houbing Song
Summary: The new generation of artificial intelligence technology has enhanced the autonomous monitoring capabilities of marine equipment. A ocean monitoring platform based on edge computing enables autonomous collaboration among multiple equipment groups. By using an improved artificial potential field method scheme, the challenges faced by multi-AUG systems operating in special underwater environments can be overcome, enabling cooperative control of the AUG group.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2023)
Article
Robotics
Christopher E. Denniston, Yun Chang, Andrzej Reinke, Kamak Ebadi, Gaurav S. Sukhatme, Luca Carlone, Benjamin Morrell, Ali-akbar Agha-mohammadi
Summary: This study describes a loop closure module that optimizes the computation of loop closures to maintain a drift-free centralized map. Experimental results demonstrate that the system can generate and maintain a map with low error, and effectively select loop closures.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Mathematical & Computational Biology
Liwei Yang, Lixia Fu, Ping Li, Jianlin Mao, Ning Guo, Linghao Du
Summary: This paper proposes a leader follower-ant colony optimization (LF-ACO) algorithm to solve the collaborative path planning problem in multi-robot systems. The algorithm incorporates new heuristic functions, leader-follower structure, and path optimization techniques to improve performance.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2022)
Article
Robotics
Michael Watterson, Sikang Liu, Ke Sun, Trey Smith, Vijay Kumar
INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH
(2020)
Article
Automation & Control Systems
Dimitra Panagou, Matthew Turpin, Vijay Kumar
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2020)
Article
Robotics
James Svacha, James Paulos, Giuseppe Loianno, Vijay Kumar
IEEE ROBOTICS AND AUTOMATION LETTERS
(2020)
Article
Robotics
Miguel Calvo-Fullana, Daniel Mox, Alexander Pyattaev, Jonathan Fink, Vijay Kumar, Alejandro Ribeiro
Summary: Multi-agent systems are crucial in modern robotics, often requiring coordination among agents through communication. ROS-NetSim serves as an interface between robotic and network simulators, offering a lightweight, modular, and adaptive approach to accurately simulate interactions and loops, regardless of the specific network or physics simulator being used. By providing transparency, independency, and tunability, ROS-NetSim enables high-fidelity representations of robotic and network interactions.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Article
Robotics
Ian D. Miller, Anthony Cowley, Ravi Konkimalla, Shreyas S. Shivakumar, Ty Nguyen, Trey Smith, Camillo Jose Taylor, Vijay Kumar
Summary: The paper introduces a real-time method for globally localizing a robot using semantics, which utilizes egocentric 3D semantically labelled LiDAR and IMU as well as top-down RGB images obtained from satellites or aerial robots. The method builds a globally registered, semantic map of the environment as it runs, showing better than 10 m accuracy, high robustness, and the ability to estimate the scale of a top-down map on the fly if it is initially unknown.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Article
Robotics
Jake Welde, James Paulos, Vijay Kumar
Summary: This research addresses the problem of planning dynamically feasible trajectories for underactuated aerial manipulators and develops a method to determine a family of flat output trajectories to precisely produce any desired task trajectory. Criteria on the manipulator geometry to ensure certain stability properties are also provided. The approach is demonstrated in simulation for systems of varying geometry and number of joints, allowing tasks to be performed dynamically without sacrificing accuracy.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Article
Robotics
Guilherme Nardari, Avraham Cohen, Steven W. Chen, Xu Liu, Vaibhav Arcot, Roseli A. F. Romero, Vijay Kumar
Summary: A novel descriptor based on Urquhart tessellations is presented, which is derived from the position of trees in a forest. The proposed framework demonstrates superior accuracy and robustness in loop closure detection experiments for map-merging from different flights of a UAV in a pine tree forest.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Article
Robotics
Matthew Malencia, Vijay Kumar, George Pappas, Amanda Prorok
Summary: The study focuses on fair redundant assignment of multi-agent tasks, proposing a supermodularity-based solution to the NP-hard optimization problem. The algorithm outperforms benchmarks and achieves improvements in fairness and utility in simulations on transport networks.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Article
Computer Science, Hardware & Architecture
Xu Liu, Steven W. Chen, Guilherme Nardari, Chao Qu, Fernando Cladera Ojeda, Camillo J. Taylor, Vijay Kumar
Summary: Mobile robots, such as UGVs and UAVs, are increasingly being used in precision agriculture. While UGVs have larger payload capabilities and longer operation time, UAVs are better suited for tasks that require fast coverage and navigation through harsh terrain. However, developing a reliable fully autonomous UAV system for extracting actionable information in large-scale cluttered agricultural environments remains challenging.
Article
Robotics
Daniel Mox, Vijay Kumar, Alejandro Ribeiro
Summary: The research team proposed a data-driven approach using convolutional neural networks to optimize the algebraic connectivity of robot teams. This method has the potential for online applications and is significantly faster than traditional optimization-based approaches.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Robotics
Xu Liu, Guilherme Nardari, Fernando Cladera Ojeda, Yuezhan Tao, Alex Zhou, Thomas Donnelly, Chao Qu, Steven W. Chen, Roseli A. F. Romero, Camillo J. Taylor, Vijay Kumar
Summary: In this letter, an integrated system for large-scale autonomous flights and real-time semantic mapping in challenging under-canopy environments is proposed. The system utilizes LiDAR data to detect and model tree trunks and ground planes, and employs a multi-level planning and mapping framework to compute dynamically feasible trajectories. This leads to the construction of a semantic map of the user-defined region of interest, while minimizing odometry drift through a drift-compensation mechanism.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Robotics
Siddharth Mayya, Ragesh K. Ramachandran, Lifeng Zhou, Vinay Senthil, Dinesh Thakur, Gaurav S. Sukhatme, Vijay Kumar
Summary: This study explores a scenario where a team of robots tracks targets or hazards while considering the possibility of sensory failures. The proposed control framework balances the competing objectives of tracking quality maximization and sensor preservation, incorporating risk-aware and adaptive components. By explicitly considering the heterogeneous sensing capabilities of the robots, a more effective risk vs. tracking quality trade-off is achieved.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Robotics
Ian D. Miller, Fernando Cladera, Trey Smith, Camillo Jose Taylor, Vijay Kumar
Summary: In this study, we propose an end-to-end heterogeneous multi-robot system framework that enables ground robots to localize, plan, and navigate in a real-time semantic map created by a high-altitude quadrotor. The ground robots autonomously choose and resolve targets, and perform cross-view localization with the overhead map. The entire system operates with an opportunistic and distributed communication backbone, requiring no external infrastructure aside from GPS for the quadrotor.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Robotics
Lifeng Zhou, Vijay Kumar
Summary: This article addresses the problem of multi-robot target tracking in adversarial environments where attacks or failures may disable robots' sensors and communications. The authors propose a robust framework that accounts for worst-case sensing and communication attacks and design a robust planning algorithm, RATT, which approximates communication attacks to sensing attacks and optimizes against them. They provide provable suboptimality bounds for the tracking quality and demonstrate RATT's effectiveness and robustness through evaluations.
IEEE TRANSACTIONS ON ROBOTICS
(2023)
Article
Robotics
Steven W. Chen, Guilherme Nardari, Elijah S. Lee, Chao Qu, Xu Liu, Roseli Ap Francelin Romero, Vijay Kumar
IEEE ROBOTICS AND AUTOMATION LETTERS
(2020)
