Article
Engineering, Electrical & Electronic
Zhishuo Li, Yunong Tian, Guodong Yang, En Li, Yanfeng Zhang, Minghao Chen, Zize Liang, Min Tan
Summary: In recent years, various types of inspection robots have been developed to automate powerline inspection. The hybrid robot combines climbing and flying abilities and shows great potential in powerline inspection. However, landing a hybrid robot on a specific powerline among multiple ones is challenging. We propose a complete solution for autonomous landing, including powerline detection, depth estimation, tracking, and landing strategies. Experimental results demonstrate the accuracy of our algorithms and the effectiveness of the robot in tracking and landing tasks.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Computer Science, Information Systems
Yang Yang, Desheng Liu
Summary: This paper proposes a distributed imaging satellite mission planning method (DISMPA) based on multi-agent system. It constructs a distributed imaging satellite mission planning model with variable collaborative division of labor and defines the intelligence level of satellites in the satellite cluster and the interaction mode between satellites. The cooperation mechanism between satellite agents is established based on blackboard model, and a hybrid discrete multi-verse optimization algorithm is proposed to solve the mission planning problem of worker agents in the model.
Article
Automation & Control Systems
Eric Bernd Gil, Genaina Nunes Rodrigues, Patrizio Pelliccione, Radu Calinescu
Summary: This paper introduces a framework called MutRoSe for Multi-Robot mission Specification and decomposition, which simplifies and automates the process of allocating concrete tasks to each robot in a multi-robot system. MutRoSe allows mission designers to define mission and environment aspects in a high-level specification language, taking into account real-world scenarios, task dependencies, and task library reusability. Additionally, MutRoSe automates the decomposition of MRS missions into task instances and allocates them to specific robots with appropriate consideration of task dependencies. The effectiveness of MutRoSe is demonstrated through the application to four missions from a published repository of MRS applications.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2023)
Article
Robotics
Graeme Best, Rohit Garg, John Keller, Geoffrey A. Hollinger, Sebastian Scherer
Summary: We present a coordinated autonomy pipeline for multi-sensor exploration of confined environments. Our solution addresses four challenges that are often overlooked in prior work and utilizes a behavior tree architecture to adaptively switch between exploration and response behaviors. The pipeline is evaluated through extensive field experiments, demonstrating its effectiveness in various environments and resilience to adverse events.
INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH
(2023)
Article
Computer Science, Information Systems
Branko Miloradovic, Baran Curuklu, Mikael Ekstrom, Alessandro Vittorio Papadopoulos
Summary: Multi-agent systems have attracted significant attention in research and industry, especially in the fields of robotics and computer science. The problem of task allocation to agents, known as Multi-Robot Task Allocation (MRTA) problem, is fundamental in robotics. Our research focuses on the neglected aspect of Multi-Task (MT) robots, and we propose two models to formalize the problem and introduce the distinction between physical and virtual tasks. We also conduct a comprehensive performance analysis of the models using CPLEX and CP Optimizer.
Article
Robotics
Baiyu Li, Hang Ma
Summary: We propose the Double-Deck Multi-Agent Pickup and Delivery (DD-MAPD) problem formulation to handle the shelf rearrangement problem in automated warehouses. We prove the NP-hardness of solving DD-MAPD and introduce the MAPF-DECOMP algorithmic framework to tackle it. Experimental results show that MAPF-DECOMP is efficient and effective in computing high-quality solutions for large-scale instances of DD-MAPD.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2023)
Article
Automation & Control Systems
Troy Bruggemann
Summary: This article proposes an automated feature-driven flight planning method to address the problems of automatically generating and optimizing flight paths for inspecting large linear infrastructure assets. Results show that automated planning can reduce human workload and enable cost-saving benefits of automated aircraft deployment.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Yinhua Liu, Wenzheng Zhao, Tim Lutz, Xiaowei Yue
Summary: This paper proposes a novel task allocation methodology for coordinated motion planning of multi-robot inspection, achieving efficient and well-balanced measurement assignment among robots. Collision-free path planning and coordinated motion planning are developed through dynamic searching and perturbation of probe poses or paths in conflicting robots. The proposed approach significantly and consistently reduces the risk of collisions, resolves conflicts among robots, and decreases inspection cycle time.
JOURNAL OF INTELLIGENT MANUFACTURING
(2022)
Article
Computer Science, Theory & Methods
Fatih Semiz, Faruk Polat
Summary: The paper introduces the Incremental Multi-Agent Path Finding (I-MAPF) problem and a solution method called CBS-D*-lite, which aims to improve efficiency by avoiding re-planning for agents not affected by environmental changes.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Computer Science, Information Systems
Eran Gefen, David Zarrouk
Summary: This paper presents an improved design of the flying-driving robot, FSTAR2, and a near optimal energy-based algorithm that allows efficient navigation in obstacle-filled environments. The experiments show significant reduction in energy consumption and higher driving speeds.
Article
Computer Science, Information Systems
Hamza Chakraa, Edouard Leclercq, Francois Guerin, Dimitri Lefebvre
Summary: This paper addresses the problem of optimally assigning a set of tasks to a set of mobile robots equipped with different sensors. A centralized Genetic Algorithm (GA) is proposed to determine the task each robot will perform. The results demonstrate that the GA approach offers a favorable balance between optimality and execution time.
Article
Computer Science, Artificial Intelligence
Shangding Gu, Jakub Grudzien Kuba, Yuanpei Chen, Yali Du, Long Yang, Alois Knoll, Yaodong Yang
Summary: The study investigates safe multi-agent reinforcement learning for multi-robot control and proposes theoretical solutions and benchmark environments for this problem.
ARTIFICIAL INTELLIGENCE
(2023)
Article
Robotics
Rahul Kala
Summary: Mission planning requires robots to solve complex tasks specified using specific languages. The paper proposes an evolutionary framework for incrementally solving the problem of mission planning, integrating task solutions through Dynamic Programming, and outperforming other methods.
INTELLIGENT SERVICE ROBOTICS
(2021)
Article
Engineering, Electrical & Electronic
Hongcan Guan, Xiliang Sun, Yanjun Su, Tianyu Hu, Haitao Wang, Heping Wang, Chigang Peng, Qinghua Guo
Summary: This paper discusses the increasing demand for electricity in recent decades, which has put pressure on powerline systems to ensure uninterrupted power supply. The use of unmanned aerial vehicles (UAVs) equipped with sensors for powerline inspections is introduced as a new technology. Additionally, an intelligent automatic inspection solution is proposed to improve inspection efficiency and reduce costs.
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS
(2021)
Article
Automation & Control Systems
Wenzheng Zhao, Yinhua Liu, Yanzheng Li, Chengwei Hu, Rui Sun
Summary: The multi-robot coverage path planning problem (MRCPP) is a highly coupled problem with complex engineering constraints. Most research in this field focuses on viewpoint sampling and sequential path planning for individual robots. However, measurement uncertainty requirements for the inspected surface features are often neglected. To address this, a MRCPP framework for free-form surface inspection is developed, considering measurement uncertainty requirements and full coverage of the surfaces.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2023)
