Xiaodong Yi

Communication-Motion Planning for Wireless Relay-Assisted Multi-Robot System

This letter focuses on a scenario in which a team of sensing robots survey an area with predefined routes, and transmit the monitored information to a remote base station through a mobile relay. In this scenario, automatically adjusting the position of the mobile relay for maintaining wireless link quality while the sensing robots are moving is a challenging problem. In this letter, we consider the problem of minimizing the total energy consumption. We propose using dynamic programming (DP) and single-step optimization. By comparing the pros and cons of both methods, we propose a novel approximate optimal communication-motion planning (AOCMP) method based on approximate dynamic programming (ADP). Simulation results demonstrate that AOCMP may sharply decrease the computation time compared with DP, while performs better than single-step optimization which proves that AOCMP achieves a beneficial energy-complexity tradeoff for solving high-dimension problems.

Distributed control for flocking and group maneuvering of nonholonomic agents

In this paper, we propose a distributed control approach for flocking and group maneuvering of nonholonomic agents, with constrained kinematic properties commonly found in practical systems, such as fixed‐wing unmanned aerial vehicles. Flocking of agents with differential drive kinematics is addressed by introducing a virtual leader–follower mechanism into the Olfati‐Sabers algorithm, which is originally proposed for holonomic agents with double integrator kinematics. Then, group maneuverability of the flock is achieved by superimposing a group motion onto each agents flocking motion. Moreover, it is proven that speed limits are intrinsically guaranteed by the approach, which renders it more applicable in practical systems. Experimental results in MATLAB and Gazebo, a popular robotic simulator, are presented to evaluate the performance and demonstrate the effectiveness of the proposed approach.

ALLIANCE-ROS: a software architecture on ROS for fault-tolerant cooperative multi-robot systems

Programming multi-robot systems is a complicated and time-consuming work, due to two challenges, i.e., the distributed multi-robot cooperation and the robot software reusability. ALLIANCE [1] is a fully distributed, fault-tolerant and behavior-based model. ROS (Robot Operating System) provides abundant robot software modules. In this paper, by combining both, we propose a software architecture named ALLIANCE-ROS for developing fault-tolerant cooperative multi-robot systems with a lot of software resources available. We encapsulate the ROS mechanisms and Python libraries to construct the basic function units of the ALLIANCE model. One may inherit them to construct the ALLIANCE-model-application with all ROS algorithms, modules and resources available. This work is demonstrated by an experiment of multi-robot patrol in both the simulated and the real environments.

Sequence searching with CNN features for robust and fast visual place recognition

Abstract The primary purpose of this paper is to realize robust place recognition algorithms towards simultaneous viewpoint and condition changes, and provide satisfactory computational efficiency. In this paper, we significantly improve the viewpoint invariance of the SeqSLAM algorithm by using state-of-the-art deep learning techniques to generate robust feature representations of images and develop the SeqCNNSLAM. Experimental results show that SeqCNNSLAM outperforms state-of-the-art place recognition systems in most cases, such as, when precision is maintained at 100%, the maximum recall obtained by SeqCNNSLAM is 50% higher than SeqSLAM on the Norland dataset with simultaneous condition change and 12.5% viewpoint change. Besides, we develop an acceleration method called A-SeqCNNSLAM, which exploits the location relationship between the matching images of adjacent images to reduce the matching range of the current image. Experimental results demonstrate that an acceleration of  ∼  5 times is achieved with minimal accuracy degradation of  ∼  5%. Finally, to enable A-SeqCNNSLAM adaptability in new environments, O-SeqCNNSLAM is devised for the online parameter adjustment in A-SeqCNNSLAM.

An interactive computer-based simulation system for endovascular aneurysm repair surgeries

This paper presents an interactive simulation system for surgical procedures of endovascular aneurysm repair. It extracts anatomical structure of clinic interest from patient‐specific X‐ray computed tomography or magnetic resonance imaging data by image segmentation techniques, and then reconstructs surface triangular meshes of these anatomical structures from the volumetric data. The core of the system is an interactive computer‐based simulation module. It consists of a physical modeling unit, a collision detection unit, a visualization unit, and a control unit. The integration of these units together makes it possible for users to interact with the system in real time, performing virtual catheterization, angiography, and stent graft deployment under a user‐specified rendering mode. The prototype system can be used as a cost‐efficient tool for surgical planning with patient‐specific anatomical geometry and for practice of surgical procedures before actual operation. Copyright

Distributed coordination with connectivity maintenance for nonholonomic robots: Robot Coordination with Connectivity Maintenance

Multirobot systems have been studied extensively in the recent years. Maintaining connectivity has significant impacts on the stability and convergence of the multirobot systems. In this work, we design a three‐layer framework for multirobot coordination. Furthermore, a novel distributed algorithm is proposed to achieve the navigation objective while satisfying connectivity maintenance and collision avoidance constraints. The algorithm is a hybrid of an rapidly exploring random tree‐based planner and an extended distributed navigation function‐based controller. The coordination framework and the distributed algorithm are demonstrated to be effective through a series of illustrative simulations. They outperform the current state‐of‐the‐art method in terms of efficiency and applicability.

Real-time Simulation of Catheterization in Endovascular Surgeries

This paper proposes an efficient and stable numerical method for modeling catheterization during endovascular surgeries. The guidewire‐catheter combination is treated as an elastic rod, which has very large resistance against twisting about its medial axis. A torsion free assumption is made, and the physical behavior of the rod is predominately governed by stretching and bending energies. This simplification greatly reduces the computational complexity and makes the model more stable, while the simulation results are still realistic enough for its application in endovascular surgeries. A contact handling algorithm that directly makes use of the volume data of the relevant tissues is proposed to simulate the interaction between the guidewire‐catheter combination and the aortic wall. During each simulation step, the penetration depth of each vertex in contact with the aortic wall is calculated using moving least squares surfaces, and the contact is then resolved in a position‐based manner. A comprehensive quantitative evaluation of the rod model is performed to validate its accuracy. Finally, the proposed approach is applied in a prototype system for simulation of endovascular aneurysm repair surgeries. Its efficiency and effectiveness are demonstrated in a real‐time interactive catheterization simulation. Copyright

Visual SLAM using multiple RGB-D cameras

In this paper, we present a solution to visual simultaneous localization and mapping (SLAM) using multiple RGB-D cameras. In the SLAM system, we integrate visual and depth measurements from those RGB-D cameras to achieve more robust pose tracking and more detailed environmental mapping in unknown environments. We present the mathematical analysis of the iterative optimizations for pose tracking and map refinement of a RGB-D SLAM system in multi-camera cases. The resulted SLAM system allows configurations of multiple RGB-D cameras with non-overlapping fields of view (FOVs). Furthermore, we provide a SLAM-based semiautomatic method for extrinsic calibration among such cameras. Finally, the experiments in complex indoor scenarios demonstrate the efficiency of the proposed visual SLAM algorithm.

CORB-SLAM: A Collaborative Visual SLAM System for Multiple Robots

With the single-robot visual SLAM method reaching maturity, the issue of collaboratively exploring unknown environments by multiple robots attracts increasing attention. In this paper, we present CORB-SLAM, a novel collaborative multi-robot visual SLAM system providing map fusing and map sharing capabilities. Experimental results on popular public datasets demonstrate the performance of the CORB-SLAM. Furthermore, we make the source code of CORB-SLAM to be publicly available (https://github.com/lifunudt/CORB-SLAM.git).

Distributed coordination with connectivity maintenance for nonholonomic robots

Wanrong Huang1 Yanzhen Wang2 Xiaodong Yi2 Xue-Jun Yang3 1State Key Laboratory of High Performance Computing, College of Computer, National University of Defense Technology, Changsha, China 2Artificial Intelligence Research Center, National Innovation Institute of Defense Technology, Beijing, China 3National Innovation Institute of Defense Technology, Beijing, China Subsequent to publication, the affiliation and citation of the article by Huang et al.1 have been modified. The correct order is presented above.