Current Issue

20 March 2019, Volume 1 Issue 1

   Next Issue

Comprehensive Reviews

The new era of artificial intelligence

Nanning ZHENG

2019, 1(1):  1-3.  doi:10.11959/j.issn.2096-6652.201914

Asbtract ( 9892 )   HTML ( 12630)   PDF (506KB) ( 9252 )   Knowledge map   

References | Related Articles | Metrics

The goal of artificial intelligence (AI) is to make machines learn,think and understand like human beings,with the focus of using computers to imitate human intelligence.It integrates knowledge of multi-disciplines including intuitive reasoning (with all kinds of physical and social common senses),computer vision,natural language perception and interaction,and machine learning.Hybrid-augmented intelligence is the typical feature of the next generation of AI,and cognitive computing of visual and auditory information is the core research content.Therefore,comprehending the human visual and auditory cognition mechanism and building its computable model is of vital significance for AI’s research and development.

Artificial intelligence is entering the post deep-learning era

Bo ZHANG

2019, 1(1):  4-6.  doi:10.11959/j.issn.2096-6652.201913

Asbtract ( 2859 )   HTML ( 1765)   PDF (494KB) ( 8214 )   Knowledge map   

References | Related Articles | Metrics

Points of views on the concept of artificial intelligence (AI),the progress AI has made,and the trend of AI’s future development were presented in this paper.

Papers

Social energy:mining energy from the society

Jun ZHANG,Fei-yue WANG,Zhou FANG

2019, 1(1):  7-20.  doi:10.11959/j.issn.2096-6652.201901

Asbtract ( 2007 )   HTML ( 45)   PDF (19145KB) ( 655 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The inherent nature of energy,i.e.,physicality,sociality and informatization,implies the inevitable and intensive interaction between energy systems and social systems.In this paper,the nature of energy was proposed,and then the concept and intention of social energy systems for electrical power were proposed.A general methodology of establishing and investigating social energy was proposed.A case study on the University of Denver (DU) campus grid was provided and studied to demonstrate the social energy concept.In the concluding remarks,the technical pathway was discussed in both social and nature sciences,to social energy,and our vision on its future.

Parallel transportation:virtual-real interaction for intelligent traffic management and control

Yisheng LV,Yuanyuan CHEN,Junchen JIN,Zhenjiang LI,Peijun YE,Fenghua ZHU

2019, 1(1):  21-33.  doi:10.11959/j.issn.2096-6652.201908

Asbtract ( 1817 )   HTML ( 138)   PDF (6986KB) ( 976 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The concept,architecture,methods and applications of parallel transportation systems were introduced.The main idea is to build software defined artificial transportation systems for specific traffic application tasks and scenarios;use computational experimental methods for experiments,analysis,evaluation,prediction,learning,and optimization;and parallel execution is conducted for management and control of urban transportation systems,leading to virtual-real interactive parallel intelligence.Parallel traffic management and control is an extension and application of ACP approach,which emphasizes using virtual-real interactive ways to generate parallel intelligence for urban transportation systems,towards improving the intelligence of urban traffic management and control.

Parallel loading and unloading:smart technology toward intelligent logistics

Dayong SHEN,Xiao WANG,Sheng LIU

2019, 1(1):  34-39.  doi:10.11959/j.issn.2096-6652.201911

Asbtract ( 1398 )   HTML ( 79)   PDF (2472KB) ( 818 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The ACP theory plays an important role in modeling and controlling complex,uncertain systems.Parallel loading and unloading is the integration of this theory in intelligent logistics.The basic framework of the parallel loading and unloading system was proposed in this paper.In this framework,the artificial system was built and updated continuously to describe real loading and unloading process,then computing experiments perform all kinds experiments for predicting all possible development and evaluate the best scheme.Finally,a bridge between the real system and the artificial system was built by the parallel execution and a long-term guidance for the real system could be realized,while artificial system was updated based on information from the real system.

Research on digital quadruplets in cyber-physical-social space-based parallel driving

Teng LIU, Xiao WANG, Yang XING, Yu GAO, Bin TIAN, Long CHEN

2019, 1(1):  40-51.  doi:10.11959/j.issn.2096-6652.201902

Asbtract ( 2130 )   HTML ( 73)   PDF (26943KB) ( 734 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Parallel driving is an advanced,cloudy and unified approach for connected automated driving that includes operation management,online condition monitoring and emergency take-over.The digital quadruplets in parallel driving,which include physical vehicle,descriptive vehicle,predictive vehicle and prescriptive vehicle were defined.First,the content and framework of the parallel driving and digital quadruplets were introduced.Then,the details of the descriptive,predictive and prescriptive vehicles were given.Finally,the applications of the digital quadruplets in mining and logistics scenarios were discussed.

Dynamical modeling and intelligent control of space soft manipulator

Xiang ZHANG,Hongwei LIU,Zhuoqun LIU,Zhenguo YAN,Xiaoqian CHEN,Yiyong HUANG

2019, 1(1):  52-61.  doi:10.11959/j.issn.2096-6652.201905

Asbtract ( 1320 )   HTML ( 44)   PDF (5770KB) ( 1004 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Space intelligent soft manipulator shows a broad prospect in the application of on-orbit service,due to its lightweight,dexterity and foldability.In this paper,a design of a modular soft manipulator with omnidirectional drivers was proposed.The dynamical model of the proposed soft manipulator was developed based on the rigid-flexible coupling system dynamics.The detection and recognition algorithm for space non-cooperative objects was designed based on deep regression convolution neural network.Both the dynamical model and the algorithm are the foundation for the intelligent control of the soft manipulator.Moreover,an experiment system of the soft manipulator was designed and developed,and the autonomous motion control and the recognition algorithm were validated.

A dynamic object tracking system of industrial robots based on visual servoing technology

Yichao MAO,Qi LU

2019, 1(1):  62-69.  doi:10.11959/j.issn.2096-6652.201903

Asbtract ( 1229 )   HTML ( 37)   PDF (7025KB) ( 386 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Visual servoing technology has high potential in many industrial applications.Conventionally,precise servoing is often achieved by using high-resolution cameras which are usually costly and unaffordable in many of industrial scenarios.A novel visual servoing control scheme which generates robot control signal based on the fusion of visual feedback and robot motion feedback was presented in the paper.By doing so,the robot control loop runs at a frequency independent to the visual feedback frequency (i.e.the frame rate of a camera).Therefore,it becomes feasible to achieve precise and dynamic servoing even with a low frame rate and low-cost camera.The new visual servoing scheme was introduced briefly.The feasibility of the scheme as well as its performances in low frame rate conditions was also illustrated experimentally.

Three-dimensional reconstruction cloud studio based on semi-supervised generative adversarial networks

Chong YU

2019, 1(1):  70-82.  doi:10.11959/j.issn.2096-6652.201909

Asbtract ( 1479 )   HTML ( 43)   PDF (12802KB) ( 928 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Because of the intrinsic complexity in computation,three-dimensional (3D) reconstruction is an essential and challenging topic in computer vision research and applications.The existing methods for 3D reconstruction often produce holes,distortions and obscure parts in the reconstructed 3D models.While the 3D reconstruction algorithms based on machine learning can only reconstruct voxelized 3D models for simple isolated objects,they are not adequate for real usage.From 2014,the generative adversarial network (GAN) is widely used in generating unreal dataset and semi-supervised learning.So the focus of this paper is to achieve high quality 3D reconstruction performance by adopting GAN principle.A novel semi-supervised 3D reconstruction framework,namely SS-GAN-3D was proposed,which can iteratively improve any raw 3D reconstruction models by training the GAN models to converge.This new model only takes 2D observation images as the weak supervision,and doesn’t rely on prior knowledge of shape models or any referenced observations.Finally,through qualitative and quantitative experiments and analysis,this new method shows compelling advantages over the current state-of-the-art methods on Tanks ＆amp; Temples and ETH3D reconstruction benchmark datasets.Based on SS-GAN-3D,the 3D reconstruction studio solution was proposed.

Formation control of mobile robots with UWB localization technology

Yajun ZHENG,Lei XUE,Lu DONG,Qingling WANG

2019, 1(1):  83-87.  doi:10.11959/j.issn.2096-6652.201907

Asbtract ( 1755 )   HTML ( 64)   PDF (1463KB) ( 1275 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In this paper,a novel multi-mobile robots formation control algorithm based on ultra-wideband (UWB) location technology was proposed.UWB location technology is relatively mature but is less applied to formation control problem.The framework of systems and hardware design of mobile multi-robot systems were first presented.Then,the systems model was established and the corresponding distributed control algorithm was proposed.Finally,the stability of multi-robot systems was proved and one simulation example was given to show the effectiveness of the proposed formation control algorithm.

Algorithm design for food-picking combining deep learning and biometrics recognition

Niya CHEN,Jiayang RUAN,Jinmiao HUANG,Wei YANG

2019, 1(1):  88-95.  doi:10.11959/j.issn.2096-6652.201906

Asbtract ( 1341 )   HTML ( 38)   PDF (6028KB) ( 693 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Irregular-shape food picking and processing is a common problem in industrial automation,which can be difficult for classical image processing techniques,because of the big variations in food shape and characteristics,also the high-performance requirements in both algorithm accuracy and speed.In this paper,a hybrid method based on deep learning and feature recognition was proposed,which first roughly localized the target point based on deep learning model,and then created a search range accordingly.After that,based on biological feature analysis,target point could be accurately localized in the search range.Based on the data of shrimps,a kind of common food,the performance of the proposed method was tested.The shrimp images were pre-processed and used to train the deep learning model for rough localization.Then the shrimp body pose was normalized for edge extraction after proper rotation and projection.The extracted edge curve in search range was analyzed to accurately localize the target joint point.The validation results based on a test set including 1000 samples prove the feasibility of the method – the final detection rate of the proposed hybrid method is 97%.The performance meets industrial requirements on this case.

Design of transnational and intercontinental electricity market with blockchain and token economics

Siyuan CHEN, Yuyang BAI, Jun ZHANG, Fei-yu WANG

2019, 1(1):  96-105.  doi:10.11959/j.issn.2096-6652.201910

Asbtract ( 1829 )   HTML ( 76)   PDF (2573KB) ( 1499 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Transnational and intercontinental electricity trading (TIET) is a fundamental enabling component and efficient method to the development of Global Energy Internet (GEI),which aims to build an energy model centered on electricity consumption.The problems faced by TIET was analyzed,and a transnational and intercontinental electricity market (TIEM) based on blockchain and token economics was proposed.The TIEM was divided into five components,and each component was described and modeled respectively.A scale-up power grid based IEEE 39-nodes system was used to analyze the proposed market.The simulation results show that TIEM based on blockchain and token economy can ensure the effective implementation of cross-regional and multi-agent electricity transactions,and also encourage the generation companies (GENCOs) to participate in electricity market with clean energy.