Pub Date : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669080
Shuhei Kawaguchi, Y. Fukuyama
This paper proposes parallel hybrid particle swarm optimization (PHPSO) for the integration framework of optimal operational planning problem of an energy plant and production scheduling problem for actual reduction of the secondary energy costs in factories. Conventionally, fixed loads of the various tertiary energies have been utilized for solving optimal operational planning of the energy plant so far. On the contrary, in this paper, the loads of the various tertiary energies are calculated according to candidates of production scheduling and actual reduction of the secondary energy costs in factories is realized. The proposed method is applied to 10 jobs and 10 machines problem and it is verified that it can minimize the secondary energy cost and production time simultaneously with higher quality solutions compared with the conventional HPSO, and realize fast computation by parallel computation using PHPSO.
{"title":"Parallel Hybrid Particle Swarm Optimization for Integration Framework of Optimal Operational Planning Problem of an Energy Plant and Production Scheduling Problem","authors":"Shuhei Kawaguchi, Y. Fukuyama","doi":"10.1109/ICAIIC.2019.8669080","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669080","url":null,"abstract":"This paper proposes parallel hybrid particle swarm optimization (PHPSO) for the integration framework of optimal operational planning problem of an energy plant and production scheduling problem for actual reduction of the secondary energy costs in factories. Conventionally, fixed loads of the various tertiary energies have been utilized for solving optimal operational planning of the energy plant so far. On the contrary, in this paper, the loads of the various tertiary energies are calculated according to candidates of production scheduling and actual reduction of the secondary energy costs in factories is realized. The proposed method is applied to 10 jobs and 10 machines problem and it is verified that it can minimize the secondary energy cost and production time simultaneously with higher quality solutions compared with the conventional HPSO, and realize fast computation by parallel computation using PHPSO.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129421338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-02-01DOI: 10.1109/ICAIIC.2019.8668976
Alberto Fornaser, M. Cecco, Teruhiro Mizumoto, K. Yasumoto
Recognition of activity of daily living (ADL) with ubiquitous sensors has been studied so far, aiming to provide services like automatic life logging, elderly monitoring and energy saving in domestic environments. Although existing studies achieve good accuracy of ADL recognition on average, mis-classification of some activities often occur. In this paper, we try to minimize mis-classification in ADL recognition through reliability assessment of the recognition results obtained by machine learning. Specifically, we propose a novel ADL recognition model which extends the random forest classifier trained by ADL data-set by adding the real time uncertainty propagation of the measured variables to each decision tree providing thus the confidence probability of each output class. This adds to the classifier output a confidence value that holds an important role for many purposes such as decision making, features design to improve the classification rate for some classes, etc. The proposed model classifies the input data samples into activity classes with high confidence probability (e.g., more than 50% confidence) and an unclassifiable class, where higher confidence probability leads to the higher recognition accuracy but higher ratio of unclassifiable samples. Through experiments, we confirmed that the proposed model achieve 75% accuracy with less than 30% unclassifiable samples and 95% accuracy with 50% unclassifiable samples.
{"title":"Reliability assessment on human activity recognition","authors":"Alberto Fornaser, M. Cecco, Teruhiro Mizumoto, K. Yasumoto","doi":"10.1109/ICAIIC.2019.8668976","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8668976","url":null,"abstract":"Recognition of activity of daily living (ADL) with ubiquitous sensors has been studied so far, aiming to provide services like automatic life logging, elderly monitoring and energy saving in domestic environments. Although existing studies achieve good accuracy of ADL recognition on average, mis-classification of some activities often occur. In this paper, we try to minimize mis-classification in ADL recognition through reliability assessment of the recognition results obtained by machine learning. Specifically, we propose a novel ADL recognition model which extends the random forest classifier trained by ADL data-set by adding the real time uncertainty propagation of the measured variables to each decision tree providing thus the confidence probability of each output class. This adds to the classifier output a confidence value that holds an important role for many purposes such as decision making, features design to improve the classification rate for some classes, etc. The proposed model classifies the input data samples into activity classes with high confidence probability (e.g., more than 50% confidence) and an unclassifiable class, where higher confidence probability leads to the higher recognition accuracy but higher ratio of unclassifiable samples. Through experiments, we confirmed that the proposed model achieve 75% accuracy with less than 30% unclassifiable samples and 95% accuracy with 50% unclassifiable samples.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124613258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-02-01DOI: 10.1109/ICAIIC.2019.8668844
Hang Liu, Xu Zhu, T. Fujii
Spectrum sensing of orthogonal frequency division multiplex (OFDM) system has always been a challenge in cognitive radios (CR). In this paper on the basis of “classification converted sensing” scheme, the cyclostationary periodogram generated by OFDM pilots is deduced in the form of images. These images are then plugged into the convolutional neural networks (CNNs) for classifications due to CNN's strength in image classification. More importantly, certain of concerns about CNN adoption in CR system is settled. Firstly, to achieve spectrum sensing against severe noise pollution and channel fading, we use the adversarial training where a CR-specific, modified training database is proposed. Then, to settle the serviceability which is constrained by the computing power at the CR user end, the input images and the CNN architecture are refined to guarantee a low-complexity but high-performance sensing scheme. Simulation results proved our method possesses an excellent sensing capability while achieving higher detection accuracy over the conventional way.
{"title":"Adversarial training for low-complexity convolutional neural networks using in spectrum sensing","authors":"Hang Liu, Xu Zhu, T. Fujii","doi":"10.1109/ICAIIC.2019.8668844","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8668844","url":null,"abstract":"Spectrum sensing of orthogonal frequency division multiplex (OFDM) system has always been a challenge in cognitive radios (CR). In this paper on the basis of “classification converted sensing” scheme, the cyclostationary periodogram generated by OFDM pilots is deduced in the form of images. These images are then plugged into the convolutional neural networks (CNNs) for classifications due to CNN's strength in image classification. More importantly, certain of concerns about CNN adoption in CR system is settled. Firstly, to achieve spectrum sensing against severe noise pollution and channel fading, we use the adversarial training where a CR-specific, modified training database is proposed. Then, to settle the serviceability which is constrained by the computing power at the CR user end, the input images and the CNN architecture are refined to guarantee a low-complexity but high-performance sensing scheme. Simulation results proved our method possesses an excellent sensing capability while achieving higher detection accuracy over the conventional way.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130633466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669084
Hyun-Woo Kim, Keonsoo Lee, Chanki Moon, Yunyoung Nam
In this paper, we present an implementation of a smart scale that can measure a subject’s weight, heart rate and detect atrial fibrillation (AF). For weight measurement, four load cell sensors are used. For measuring heart rates and detecting AF, PSL-iECG2 is used. Load cell sensors and PSL-iECG2 are connected to Arduino Uno. As Arduino Uno has not enough computing power to analyze ECG signals and determine AF, Arduino Uno is connected to smartphone in Bluetooth. From the ECG signals, R peaks are extracted and using the R-R intervals, heart rates are calculated. AF is detected using RMSSD and Shannon entropy extracted from R-R intervals. We evaluate three classifiers that are kNN, DT, and NNs. The accuracies of each classifier for detecting AF are 83.7%, 83.7%, and 89.1%, respectively.
{"title":"Comparative Analysis of Machine Learning Algorithms along with Classifiers for AF Detection using a Scale","authors":"Hyun-Woo Kim, Keonsoo Lee, Chanki Moon, Yunyoung Nam","doi":"10.1109/ICAIIC.2019.8669084","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669084","url":null,"abstract":"In this paper, we present an implementation of a smart scale that can measure a subject’s weight, heart rate and detect atrial fibrillation (AF). For weight measurement, four load cell sensors are used. For measuring heart rates and detecting AF, PSL-iECG2 is used. Load cell sensors and PSL-iECG2 are connected to Arduino Uno. As Arduino Uno has not enough computing power to analyze ECG signals and determine AF, Arduino Uno is connected to smartphone in Bluetooth. From the ECG signals, R peaks are extracted and using the R-R intervals, heart rates are calculated. AF is detected using RMSSD and Shannon entropy extracted from R-R intervals. We evaluate three classifiers that are kNN, DT, and NNs. The accuracies of each classifier for detecting AF are 83.7%, 83.7%, and 89.1%, respectively.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126916456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-02-01DOI: 10.1109/ICAIIC.2019.8668992
H. You, Hyung-jik Kim, Dong-Kyun Joo, Seung Min Lee, Jeongung Kim, Sunwoong Choi
Near Infrared (NIR) spectroscopy is fast and non-destructive methods for analyzing materials without pretreatment. Especially as portable NIR spectrometers have been developed, the research of spectral analysis has applied to various open environment and field. In this paper, we classify visually indistinguishable eight food powders using portable VIS-NIR spectrometer with a wavelength range of 450 to 1000 nm with CNN (Convolutional Neural Network), one of the machine learnings. Further we consider open set recognition where unknown classes should be rejected at test time. The proposed CNN model achieved an accuracy of 100% for eight food powders, and 91.2% with open set. Our experimental results demonstrate the potential of material analysis using a portable VIS-NIR spectrometer with machine learning.
{"title":"Classification of Food Powders with Open Set using Portable VIS-NIR Spectrometer","authors":"H. You, Hyung-jik Kim, Dong-Kyun Joo, Seung Min Lee, Jeongung Kim, Sunwoong Choi","doi":"10.1109/ICAIIC.2019.8668992","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8668992","url":null,"abstract":"Near Infrared (NIR) spectroscopy is fast and non-destructive methods for analyzing materials without pretreatment. Especially as portable NIR spectrometers have been developed, the research of spectral analysis has applied to various open environment and field. In this paper, we classify visually indistinguishable eight food powders using portable VIS-NIR spectrometer with a wavelength range of 450 to 1000 nm with CNN (Convolutional Neural Network), one of the machine learnings. Further we consider open set recognition where unknown classes should be rejected at test time. The proposed CNN model achieved an accuracy of 100% for eight food powders, and 91.2% with open set. Our experimental results demonstrate the potential of material analysis using a portable VIS-NIR spectrometer with machine learning.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126074230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669020
Hsinying Liang, Hao-Yue Jiang
The artificial bee colony-based SLM (ABC-SLM) scheme, which is a novel PAPR reduction scheme, has been proposed to reduce the peak to average power ratio (PAPR) in orthogonal frequency division multiplexing (OFDM) systems. High PAPR can degrade the efficiency of high-power amplifier and is one of the major disadvantages of OFDM systems. This paper proposes a modified ABC-SLM scheme to further improve the PAPR reduction performance of ABC-SLM scheme. The proposed method, called GA-ABC-SLM, is combined the gene algorithm (GA) with the artificial bee colony-based SLM scheme. The simulation results show that the GA-ABC-SLM scheme has better PAPR reduction performance than the ABC-SLM scheme.
{"title":"The Modified Artificial Bee Colony-Based SLM Scheme for PAPR Reduction in OFDM Systems","authors":"Hsinying Liang, Hao-Yue Jiang","doi":"10.1109/ICAIIC.2019.8669020","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669020","url":null,"abstract":"The artificial bee colony-based SLM (ABC-SLM) scheme, which is a novel PAPR reduction scheme, has been proposed to reduce the peak to average power ratio (PAPR) in orthogonal frequency division multiplexing (OFDM) systems. High PAPR can degrade the efficiency of high-power amplifier and is one of the major disadvantages of OFDM systems. This paper proposes a modified ABC-SLM scheme to further improve the PAPR reduction performance of ABC-SLM scheme. The proposed method, called GA-ABC-SLM, is combined the gene algorithm (GA) with the artificial bee colony-based SLM scheme. The simulation results show that the GA-ABC-SLM scheme has better PAPR reduction performance than the ABC-SLM scheme.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126463515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669012
E. Dahlman, S. Parkvall, J. Peisa, H. Tullberg, H. Murai, M. Fujioka
With the completion of the first release (Rel-15) of the 3GPP fifth-generation (5G) NR specifications [1, 2], the research community should now direct its focus towards the next step in the evolution of wireless mobile communication. Similar to earlier generations, it can be expected that the next ten years will see a gradual evolution of NR, introducing new innovative technology components and further enhancing the capabilities and expanding the scope of 5G wireless access. In a longer-time perspective, we may see the emergence of completely new “beyond 5G” radio-access technology.
{"title":"Artificial Intelligence in Future Evolution of Mobile Communication","authors":"E. Dahlman, S. Parkvall, J. Peisa, H. Tullberg, H. Murai, M. Fujioka","doi":"10.1109/ICAIIC.2019.8669012","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669012","url":null,"abstract":"With the completion of the first release (Rel-15) of the 3GPP fifth-generation (5G) NR specifications [1, 2], the research community should now direct its focus towards the next step in the evolution of wireless mobile communication. Similar to earlier generations, it can be expected that the next ten years will see a gradual evolution of NR, introducing new innovative technology components and further enhancing the capabilities and expanding the scope of 5G wireless access. In a longer-time perspective, we may see the emergence of completely new “beyond 5G” radio-access technology.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127542451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-02-01DOI: 10.1109/ICAIIC.2019.8668979
Ju-Bong Kim, Do-Hyung Kwon, Yong-Geun Hong, Hyun-kyo Lim, Min Suk Kim, Youn-Hee Han
A rotary inverted pendulum is an unstable and highly nonlinear device and is used as a common model for engineering applications in linear and nonlinear control. In this study, we created a cyber physical system (CPS) to demonstrate that a deep reinforcement learning agent using a rotary inverted pendulum can successfully control a remotely located physical device. The device we created is composed of a cyber environment and physical environment using the Message Queuing Telemetry Transport (MQTT) protocol with an Ethernet connection to connect the cyber environment and the physical environment. The reinforcement learning agent controls the physical device, which is located remotely from the controller and a classical proportional integral derivative (PID) controller is utilized to implement imitation and reinforcement learning and facilitate the learning process. In addition, the control and monitoring system is built on the open source EdgeX platform, so that learning tasks performed near the source of data generation and real-time data emitted from the physical device can be observed while reinforcement learning is performed. From our CPS experimental system, we verify that a deep reinforcement learning agent can control a remotely located real-world device successfully.
{"title":"Deep Q-Network Based Rotary Inverted Pendulum System and Its Monitoring on the EdgeX Platform","authors":"Ju-Bong Kim, Do-Hyung Kwon, Yong-Geun Hong, Hyun-kyo Lim, Min Suk Kim, Youn-Hee Han","doi":"10.1109/ICAIIC.2019.8668979","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8668979","url":null,"abstract":"A rotary inverted pendulum is an unstable and highly nonlinear device and is used as a common model for engineering applications in linear and nonlinear control. In this study, we created a cyber physical system (CPS) to demonstrate that a deep reinforcement learning agent using a rotary inverted pendulum can successfully control a remotely located physical device. The device we created is composed of a cyber environment and physical environment using the Message Queuing Telemetry Transport (MQTT) protocol with an Ethernet connection to connect the cyber environment and the physical environment. The reinforcement learning agent controls the physical device, which is located remotely from the controller and a classical proportional integral derivative (PID) controller is utilized to implement imitation and reinforcement learning and facilitate the learning process. In addition, the control and monitoring system is built on the open source EdgeX platform, so that learning tasks performed near the source of data generation and real-time data emitted from the physical device can be observed while reinforcement learning is performed. From our CPS experimental system, we verify that a deep reinforcement learning agent can control a remotely located real-world device successfully.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129133025","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The aim of this study is detecting the edge of the medical MRI (Magnetic Resonance Imaging) images. This paper describes as efficient and accurate enhancement and detection method by designing 5-Tap bandpass filter signal and two high frequency sub-band MRI images in digital wavelet domain. Simulation results shows that the proposed method has high accuracy and enhancement in detecting the edge images as is compared to existing method. It provide a helpful and efficient solution for detecting disease lots of medical MRI image, and this method provide new insights in overcoming the scale sensitivity and noises in edge detection.
{"title":"Design of Wavelet Digital Filter for Edge Detection of Medical MRI image","authors":"Woon Cho, Daewon Chung, Gyungmin Hwang, Joonhyeon Jeon","doi":"10.1109/ICAIIC.2019.8668966","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8668966","url":null,"abstract":"The aim of this study is detecting the edge of the medical MRI (Magnetic Resonance Imaging) images. This paper describes as efficient and accurate enhancement and detection method by designing 5-Tap bandpass filter signal and two high frequency sub-band MRI images in digital wavelet domain. Simulation results shows that the proposed method has high accuracy and enhancement in detecting the edge images as is compared to existing method. It provide a helpful and efficient solution for detecting disease lots of medical MRI image, and this method provide new insights in overcoming the scale sensitivity and noises in edge detection.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"164 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130988330","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669037
Juntae Kim, G. Lim, Youngi Kim, Bokyeong Kim, Changseok Bae
Recent outstanding progresses in artificial intelligence researches enable many tries to implement self-driving cars. However, in real world, there are a lot of risks and cost problems to acquire training data for self-driving artificial intelligence algorithms. This paper proposes an algorithm to collect training data from a driving game, which has quite similar environment to the real world. In the data collection scheme, the proposed algorithm gathers both driving game screen image and control key value. We employ the collected data from virtual game environment to learn a deep neural network. Experimental result for applying the virtual driving game data to drive real world children’s car show the effectiveness of the proposed algorithm.
{"title":"Deep Learning Algorithm using Virtual Environment Data for Self-driving Car","authors":"Juntae Kim, G. Lim, Youngi Kim, Bokyeong Kim, Changseok Bae","doi":"10.1109/ICAIIC.2019.8669037","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669037","url":null,"abstract":"Recent outstanding progresses in artificial intelligence researches enable many tries to implement self-driving cars. However, in real world, there are a lot of risks and cost problems to acquire training data for self-driving artificial intelligence algorithms. This paper proposes an algorithm to collect training data from a driving game, which has quite similar environment to the real world. In the data collection scheme, the proposed algorithm gathers both driving game screen image and control key value. We employ the collected data from virtual game environment to learn a deep neural network. Experimental result for applying the virtual driving game data to drive real world children’s car show the effectiveness of the proposed algorithm.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134043081","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：