scholarly journals LMD-TShip*: Vision Based Large-Scale Maritime Ship Tracking Benchmark for Autonomous Navigation Applications

IEEE Access ◽  
2021 ◽  
pp. 1-1
Author(s):  
Yunxiao Shan ◽  
Shanghua Liu ◽  
Yunfei Zhang ◽  
Min Jing ◽  
Huawei Xu
Keyword(s):  
Autonomous Navigation ◽  
Large Scale

Distributed multi-robot collision avoidance via deep reinforcement learning for navigation in complex scenarios

2020 ◽  
Vol 39 (7) ◽  
pp. 856-892 ◽  
Author(s):  
Tingxiang Fan ◽  
Pinxin Long ◽  
Wenxi Liu ◽  
Jia Pan
Keyword(s):  
Reinforcement Learning ◽  
Collision Avoidance ◽  
Autonomous Navigation ◽  
Large Scale ◽  
Learning Algorithm ◽  
Free Action ◽  
Parameter Tuning ◽  
Movement Velocity ◽  
Robot Systems ◽  
Multi Robot

Developing a safe and efficient collision-avoidance policy for multiple robots is challenging in the decentralized scenarios where each robot generates its paths with limited observation of other robots’ states and intentions. Prior distributed multi-robot collision-avoidance systems often require frequent inter-robot communication or agent-level features to plan a local collision-free action, which is not robust and computationally prohibitive. In addition, the performance of these methods is not comparable with their centralized counterparts in practice. In this article, we present a decentralized sensor-level collision-avoidance policy for multi-robot systems, which shows promising results in practical applications. In particular, our policy directly maps raw sensor measurements to an agent’s steering commands in terms of the movement velocity. As a first step toward reducing the performance gap between decentralized and centralized methods, we present a multi-scenario multi-stage training framework to learn an optimal policy. The policy is trained over a large number of robots in rich, complex environments simultaneously using a policy-gradient-based reinforcement-learning algorithm. The learning algorithm is also integrated into a hybrid control framework to further improve the policy’s robustness and effectiveness. We validate the learned sensor-level collision-3avoidance policy in a variety of simulated and real-world scenarios with thorough performance evaluations for large-scale multi-robot systems. The generalization of the learned policy is verified in a set of unseen scenarios including the navigation of a group of heterogeneous robots and a large-scale scenario with 100 robots. Although the policy is trained using simulation data only, we have successfully deployed it on physical robots with shapes and dynamics characteristics that are different from the simulated agents, in order to demonstrate the controller’s robustness against the simulation-to-real modeling error. Finally, we show that the collision-avoidance policy learned from multi-robot navigation tasks provides an excellent solution for safe and effective autonomous navigation for a single robot working in a dense real human crowd. Our learned policy enables a robot to make effective progress in a crowd without getting stuck. More importantly, the policy has been successfully deployed on different types of physical robot platforms without tedious parameter tuning. Videos are available at https://sites.google.com/view/hybridmrca .

Teaching-Playback Navigation Without a Consistent Map

2018 ◽  
Vol 30 (4) ◽  
pp. 591-597 ◽  
Author(s):  
Naoki Akai ◽  
Luis Yoichi Morales ◽  
Hiroshi Murase ◽  
◽  
Keyword(s):  
Autonomous Navigation ◽  
Large Scale ◽  
Mobile Robot Navigation ◽  
Autonomous Mobile Robot ◽  
Environment Map ◽  
Localization And Mapping ◽  
Mapping Process ◽  
Line Processes ◽  
Target Path ◽  
Navigation Method

This paper presents a teaching-playback navigation method that does not require a consistent map built using simultaneous localization and mapping (SLAM). Many open source projects related to autonomous navigation including SLAM have been made available recently; however, autonomous mobile robot navigation in large-scale environments is still difficult because it is difficult to build a consistent map. The navigation method presented in this paper uses several partial maps to represent an environment map. In other words, the complex mapping process is not necessary to begin autonomous navigation. In addition, the trajectory that the robot travels in the mapping phase can be directly used as a target path. As a result, teaching-playback autonomous navigation can be achieved without any off-line processes. We tested the navigation method using log data taken in the environment of the Tsukuba Challenge and the testing results show its performance. We provide source code for the navigation method, which includes modules required for autonomous navigation (https://github.com/NaokiAkai/AutoNavi).

Mobile robot motion control and autonomous navigation in GPS-denied outdoor environments using 3D laser scanning

2019 ◽  
Vol 39 (3) ◽  
pp. 469-478
Author(s):  
Qifeng Yang ◽  
Daokui Qu ◽  
Fang Xu ◽  
Fengshan Zou ◽  
Guojian He ◽  
...  
Keyword(s):  
Mobile Robot ◽  
Motion Control ◽  
Autonomous Navigation ◽  
Laser Scanning ◽  
Large Scale ◽  
Laser Scanner ◽  
Geometric Feature ◽  
Content Type ◽  
Localization Accuracy ◽  
Novel Approach

Purpose This paper aims to propose a series of approaches to solve the problem of the mobile robot motion control and autonomous navigation in large-scale outdoor GPS-denied environments. Design/methodology/approach Based on the model of mobile robot with two driving wheels, a controller is designed and tested in obstacle-cluttered scenes in this paper. By using the priori “topology-geometry” map constructed based on the odometer data and the online matching algorithm of 3D-laser scanning points, a novel approach of outdoor localization with 3D-laser scanner is proposed to solve the problem of poor localization accuracy in GPS-denied environments. A path planning strategy based on geometric feature analysis and priority evaluation algorithm is also adopted to ensure the safety and reliability of mobile robot’s autonomous navigation and control. Findings A series of experiments are conducted with a self-designed mobile robot platform in large-scale outdoor environments, and the experimental results show the validity and effectiveness of the proposed approach. Originality/value The problem of motion control for a differential drive mobile robot is investigated in this paper first. At the same time, a novel approach of outdoor localization with 3D-laser scanner is proposed to solve the problem of poor localization accuracy in GPS-denied environments. A path planning strategy based on geometric feature analysis and priority evaluation algorithm is also adopted to ensure the safety and reliability of mobile robot’s autonomous navigation and control.

Integrated Autonomous Navigation System and Automatic Large Scale Three Dimensional Map Construction

2015 ◽  
Vol 27 (4) ◽  
pp. 401-409 ◽  
Author(s):  
Yusuke Fujino ◽  
◽  
Kentaro Kiuchi ◽  
Shogo Shimizu ◽  
Takayuki Yokota ◽  
...  
Keyword(s):  
Navigation System ◽  
Autonomous Navigation ◽  
Large Scale ◽  
Three Dimensional ◽  
Autonomous Driving ◽  
Autonomous Mobile Robot ◽  
Mobile Objects ◽  
Map Construction ◽  
Environment Recognition ◽  
Navigation Accuracy

<div class=""abs_img""> <img src=""[disp_template_path]/JRM/abst-image/00270004/10.jpg"" width=""300"" /> Constructed large-scale 3D map</div> The method we propose for constructing a large three-dimensional (3D) map uses an autonomous mobile robot whose navigation system enables the map to be constructed. Maps are vital to autonomous navigation, but constructing and updating them while ensuring that they are accurate is challenging because the navigation system usually requires accurate maps. We propose a navigation system that explores areas not explored before. The proposed system mainly uses LIDARs for determining its own position – a process known as localization – or the environment around the robot – a process known as environment recognition – for creating local maps and for avoiding mobile objects – a process known as motion planning. We constructed a detailed 3D map automatically using autonomous driving data to improve navigation accuracy without increasing the operator’s workload, confirming the feasibility of the proposed method through experiments. </span>

How cars became mobile spatial media: A geographical political economy of on-board navigation

2019 ◽  
Vol 7 (3) ◽  
pp. 362-379 ◽  
Author(s):  
Luis F. Alvarez Leon
Keyword(s):  
Political Economy ◽  
Autonomous Navigation ◽  
Large Scale ◽  
Mobile Media ◽  
Locative Media ◽  
The Core ◽  
Informational Capitalism ◽  
Economic Forces ◽  
Theoretical Developments

While cars, by definition and necessity, have always been (auto)mobile, they are not often considered, or studied, as media. In light of this, the present article seeks to elucidate the technological, political, and economic forces that have converged to transform cars into mobile spatial media. This article provides a framework to contextualize the nodal role of navigation in creating the conditions for a large-scale disruption in automobile technologies with potentially far-reaching impacts: autonomous navigation. The arguments at the core of this article bring together, and build on, recent theoretical developments in (a) locative media, (b) automobiles as mobile media, and (c) new mapping technologies, practices, and spatial media to provide a coherent perspective of cars as mobile spatial media. Informed by a geographical political economy of navigation, this perspective contributes to our understanding of cars in the context of digital and informational capitalism as these vehicles undergo qualitative transformations catalyzed by increased digital interconnections and comprehensive automation.

scholarly journals Memetic Chains for Improving the Local Wireless Sensor Networks Localization in Urban Scenarios

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2458
Author(s):  
Paula Verde ◽  
Javier Díez-González ◽  
Rubén Ferrero-Guillén ◽  
Alberto Martínez-Gutiérrez ◽  
Hilde Perez
Keyword(s):  
Autonomous Navigation ◽  
Large Scale ◽  
Ad Hoc ◽  
Indoor Localization ◽  
Memetic Algorithms ◽  
Variable Neighborhood Descent ◽  
Positioning Systems ◽  
Node Location ◽  
Urban Scenario ◽  
Design Requirements

Local Positioning Systems (LPS) have become an active field of research in the last few years. Their application in harsh environments for high-demanded accuracy applications is allowing the development of technological activities such as autonomous navigation, indoor localization, or low-level flights in restricted environments. LPS consists of ad-hoc deployments of sensors which meets the design requirements of each activity. Among LPS, those based on temporal measurements are attracting higher interest due to their trade-off among accuracy, robustness, availability, and costs. The Time Difference of Arrival (TDOA) is extended in the literature for LPS applications and consequently we perform, in this paper, an analysis of the optimal sensor deployment of this architecture for achieving practical results. This is known as the Node Location Problem (NLP) and has been categorized as NP-Hard. Therefore, heuristic solutions such as Genetic Algorithms (GA) or Memetic Algorithms (MA) have been applied in the literature for the NLP. In this paper, we introduce an adaptation of the so-called MA-Solis Wets-Chains (MA-SW-Chains) for its application in the large-scale discrete discontinuous optimization of the NLP in urban scenarios. Our proposed algorithm MA-Variable Neighborhood Descent-Chains (MA-VND-Chains) outperforms the GA and the MA of previous proposals for the NLP, improving the accuracy achieved by 17% and by 10% respectively for the TDOA architecture in the urban scenario introduced.

Sign in / Sign up

Export Citation Format

Share Document