Research on LADRC strategy of PMSM for road-sensing simulation based on differential evolution algorithm

Zhang, Hui,Wang, Yuyuan,Zhang, Guowang,Tang, Conghui The Korean Institute of Power Electronics 2020 JOURNAL OF POWER ELECTRONICS Vol.20 No.4

A linear active disturbance rejection control (LADRC) strategy for permanent magnet synchronous motor (PMSM) for road-sensing based on the differential evolution (DE) algorithm is proposed in this paper, called DE-LADRC, to obtain the better dynamic and steady-state responses of the road-sensing simulation in electric vehicle (EV) steering-by-wire (SBW) systems. Since the novel control method ignores the time delay modules in digital motor control, the controlled object is regarded as a first-order inertial link to design a first-order LADRC controller. Then aiming to solve the problem where the values of the controller parameters are changed and difficult to tune due to ignoring the time delay modules in the first-order LADRC controller, the differential evolution (DE) algorithm is designed to find the optimal controller parameters by self-tuning. Experiment results indicate the effectiveness and convergence of the DE-LADRC, as well as the correctness of the road-sensing planning. In addition, the DE-LADRC can provide the smoother feel and real-time road-sensing for driver due to experiment of a Hardware in the Loop (HIL) platform under convergent iteration.

자동 절단과 부하 감응 제어 기술을 적용한 양날 도로절단기 개발

서명국,강명철,박종호,김영진 사단법인 유공압건설기계학회 2024 드라이브·컨트롤 Vol.21 No.1

With the recent development of intelligence and automation technologies for construction machinery, thedemand for safety and efficiency of road-cutting operations has continued to increase. In response to this, adouble-blade road cutter has been developed that can automatically cut roads. However, a double-blade road cutterhas a load difference between the two blades due to the ground and wear conditions of the cutting blades. Thedifference in load between the two blades distorts the direction of travel of the cutter. In this study, a visionsensor-based driving guide technology was developed to correct the driving path of road cutters. In addition, wedeveloped a load-sensing technology that detects blade loads in real-time and controls driving speed in the event ofoverload.

다중센싱 기술을 이용한 도로결함 특성분석

김정국(Jeongguk Kim) 대한기계학회 2018 대한기계학회 춘추학술대회 Vol.2018 No.12

In this study, the defects characterization on the road surface were investigated by using multiple sensing technology such as infrared thermal imaging and laser scanning. First, a signal analysis algorithm for scanning the road surface was made and the performance of the algorithm was evaluated. The artificial specimens were prepared, and analyzed the signal algorithms of infrared thermal imaging and laser scanning through basic experiments. Infrared thermal imaging technique reflects the temperature sensitivity of the artificial defect specimen. In addition, it was found that it is possible to detect defects of various sizes through laser scanning technology. Based on this preliminary experiment, it was found that it is possible to quantitatively evaluate the defects existing on the road in the performance evaluation on actual roads. In this paper, we introduce the results of road defects analysis.

자율주행용 자계도로의 3차원 해석 및 차량위치검출시스템

유영재(Young-Jae Ryoo) 한국지능시스템학회 2005 한국지능시스템학회논문지 Vol.15 No.1

이 논문에서는 지능형교통시스템에서 자율주행용 자계도로를 3차원적으로 해석하여 차량의 위치를 검출하기 위한 시스템에 관하여 다룬다. 특히 자율주행시스템을 구성하는 핵심요소 중 하나인 위치검출시스템의 새로운 방법을 제안한다. 기존의 위치검출시스템은 자계와 차량의 위치 관계를 맵핑하는 원리를 이용한다. 이는 데이터의 저장이 필수적이며 대용량의 메모리를 요구되어 상용화시 고비용의 문제점을 가지고 있다. 이 논문에서는 기존 위치검출시스템이 가지는 문제점을 극복하기 위한 방법으로 신경망을 이용한 위치검출시스템을 제안한다. 그리고 제안한 위치검출시스템을 적용한 자율주행시스템을 설계한다. 설계한 자율주행시스템의 적용 가능성을 파악하기 위하여 자율주행실험을 행하고 이를 분석한다. In this paper, a 3-dimensional analysis of magnetic road and a position sensing system for an autonomous vehicle system is described. Especially, a new position sensing system, one of the important component of an autonomous vehicle, is proposed. In a magnet based autonomous vehicle system, to sense the vehicle position, the sensor measures the field of magnetic road. The field depends on the sensor position of the vehicle on the magnetic road. As the relation between the magnetic field and the sensor position is highly complex, it is difficult that the relation is stored in memory. Thus, a neural network is used to learn the mapping from th field to the position. The autonomous vehicle system with the proposed position sensing system is tested in experimental setup.

Active RFID 가로등을 이용한 이동체 실시간 위치결정기법 연구

차맹규(Cha Maeng Q),장안진(Chang Ahn Jin),이재빈(Lee Jae Bin),유기윤(Yu Kiyun) 대한토목학회 2007 대한토목학회논문집 D Vol.27 No.1D

위치기반서비스 분야에서 RFID가 갖는 장점은 위치인식이 가능하다는 것이다. 본 연구의 목적은 RFID에 위치인식을 위한 기준점을 설정하고, 태그와 모바일 객체간의 신호들이 갖게 되는 시간차이로부터 거리를 측정하는 TDOA(Time Difference of Arrival) 기법을 적용하여 이동체의 위치를 결정하는 것이다. 이를 위해 본 연구에서는 위치인식 가능한 RFID 참조기준점의 역할로서 가로등을 채택하였다. 위치참조 기준점으로서 가로등이 갖는 장점은 측량성과를 마친 정확한 위치 자료가 이미 구축되어 있으며, 전원 공급이 필요한 능동형(active) 방식의 RFID 설치를 위해 가로등의 전원을 이용할 수 있다는 것이다, 본 연구에서는 가로등에 RFID 측위기준점을 적용한 시뮬레이션 실험을 통해 연구의 유효성을 확인한다. 실험과정에서는 이동체의 정확한 위치인식을 위하여 최적화된 태그의 개수와 신호 전송주기 및 이동체 속도를 고려하였다. RFID (Radio Frequency Identification) is regarded as a device to facilitate location sensing. Such advantages of RFID can be integrated with LBS (Location Based Service) achieving much synergy effects. This study propose and test a location sensing system using the active RFID tags. Since the coordinates of road ramps are already stored in a database as a digital map form and the road ramps placed along the curbs construct a sort of network, it is proposed to use these road ramps as reference points. Furthermore, these road ramps can provide electrical power to the active RFID tags. Thus, each moving object having a RFlD reader receives signals from these active tags and decodes corresponding coordinates of the road ramps. Then, the coordinates of a moving object are determined in real-time using the TDOA (Time Difference of Arrival) algorithm. To test the availability of this system, various kinds of constrained simulations are designed and are conducted. In these simulations, the number of active tags used, the transmission interval of RFID reader devices, and the speed of moving objects are considered as constrained factors.

김명인 시의 ‘길’ 이미지 전개 양상—1990년대의 경우

한원균 한국문예창작학회 2023 한국문예창작 Vol.22 No.1

Kim Myeongin’s poems in the 1990s can be evaluated as internalizing the problems of his era through the process of symbolic change of ‘road’. The perception of reality shown in the collection of poems Dongducheon(1979) has undergone a fundamental change process since the 1990s, which can be summarized as the revealing of ‘absence consciousness’ through the poetic method of ‘going on the road’. When the 1990s is viewed as a unit of recognition, it can be established from a sociological viewpoint of literature that embodies the mental emptiness of an era in which the central value has disappeared. Kim Myeongin poetizes this in three dimensions. 1) Internalization of ‘absence consciousness’ 2) ‘Loss of place’ and critical consciousness 3) Mental struggle process of ‘new discovery’. The sense of absence does not mean the loss of a specific object or relative poverty at the individual level, but reveals contemporaneity in that it symbolizes the loss and collapse of all values, and the consequent recognition of lack of place emerges as an ‘existential self-preservation need’. In this respect, it can be seen as one of critical understanding of reality. In such a situation, it is natural for the process of internal struggle to pursue a new direction to appear, but Kim Myeongin seeks for an inner ‘landscape’ and aims for mental freedom. Attempts to understand Kim Myungin’s poetry in light of the circumstances of the 1990s have the meaning of a research history of exploring the universal meaning of the consciousness of absence that his contemporaries had, and it is judged to provide a valid basis for examining the development process of Kim Myungin’s poetry. 김명인의 1990년대 시는 ‘길’의 상징적 변화과정을 통해 자기시대의 문제를 내면화한 것으로 평가할 수 있다. 시집 『동두천』(1979)에서 보여준 현실인식은 1990년대 이후 근본적인 변화과정을 거치는데 그것은 ‘길을 가다’라는 시적 방법을 통한 ‘부재의식’의 드러냄으로 요약할 수 있다. 1990년대를 하나의 인식단위로 볼 때, 그것은 중심의 가치가 사라져버린 시대의 정신적 공동감(空洞感)을 형상화하는 문학사회학적 관점으로 성립될 수 있다. 김명인은 이를 세 가지 차원에서 시화한다. 1) ‘부재의식’의 내면화 2) ‘장소상실’과 비판의식 3) 내면적 ‘풍경의 발견’을 위한 정신적 고투 과정이 그것이다. 부재의식은 특정한 대상의 상실이나 개인적 차원의 상대적 빈곤을 의미하는 것이 아니라 모든 가치의 상실과 함몰을 상징하고 있다는 점에서 동시대성을 드러내는 것이며 이에 따른 장소 결핍인식은 ‘실존적 자기보전 욕구’로 등장한다는 점에서 비판적 현실이해의 하나로 볼 수 있다. 이같은 상황에서 새로운 방향성을 추구하려는 내면적 고투의 과정이 나타나는 것은 자연스러운데 김명인은 내면의 ‘풍경’을 모색하고 정신적 자재(自在)를 지향하고자 한다. 김명인의 시를 1990년대적 상항에 비추어 이해하려는 시도는 동시대가 갖고 있었던 부재의식의 보편적 의미를 탐구한다는 연구사적 의미를 지니며 이후 김명인 시의 전개과정을 살피는 데 유효한 기초를 제공한다고 판단된다.

A critical analysis of road network extraction using remote sensing images with deep learning

Palvi Sharma,Rakesh Kumar,Meenu Gupta,Anand Nayyar 대한공간정보학회 2024 Spatial Information Research Vol.32 No.4

The Extraction of Roads from Remote Sensing Imagery is a rapidly developing field that has significant impacts on both the economic and social domains. In the fields of urban planning, transportation management, and disaster response, accurate and up-to-date road information obtained from satellite and aerial images is essential. Through an in depth-analysis of the existing research, this study identified the research gaps and proposed a framework for Road Extraction. The data for review is collected from the IEEE Xplore, Scopus and Web of Science where 2018–2023 publications are considered. To review the facts, 1198 articles are extracted based on Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines. After meeting the exclusion and inclusion criteria, 44 articles are identified for final considerations. In this study, a thorough investigation on road network model and road features in the context of Remote Sensing Images is discussed. Additionally, we identified a clear gap in the literature where these important elements have either not been thoroughly investigated or not mentioned at all. This paper contributes to the field of Road Extraction by providing accessible datasets with links for researchers. A comparative analysis of existing Deep Learning models is conducted, aiding researchers in model selection. Furthermore, limitations and challenges faced by researchers are highlighted, offering valuable insights for future work.

IRI estimation using analysis of dynamic tire pressure and axle acceleration

Yubo Zhao,J. Gregory McDaniel,Ming L. Wang 국제구조공학회 2017 Smart Structures and Systems, An International Jou Vol.19 No.2

A new method is developed to estimate road profile in order to estimate IRI based on the ASTM standard. This method utilizes an accelerometer and a Dynamic Tire Pressure Sensor (DTPS) to estimate road roughness. The accelerometer measures the vertical axle acceleration. The DTPS, which is mounted on the tire\' s valve stem, measures dynamic pressure inside the tire while driving. Calibrated transfer functions are used to estimate road profile using the signals from the two sensors. A field test was conducted on roads with different quality conditions in the city of Brockton, MA. The IRI values estimated with this new method match the actual road conditions measured with Pavement Condition Index (PCI) based on the ASTM standard, images taken from an onboard camera and passengers\' perceptions. IRI has negative correlation with PCI in general since they have overlapping features. Compared to the current method of IRI measurement, the advantage of this method is that a) the cost is reduced; b) more space is saved; c) more time is saved; and d) mounting the two sensors are universally compatible to most cars and vans. Therefore, this method has the potential to provide continuous and global monitoring the health of roadways.

A Novel Road Segmentation Technique from Orthophotos Using Deep Convolutional Autoencoders

( Maher Ibrahim Sameen ),( Biswajeet Pradhan ) 대한원격탐사학회 2017 大韓遠隔探査學會誌 Vol.33 No.4

This paper presents a deep learning-based road segmentation framework from very high-resolution orthophotos. The proposed method uses Deep Convolutional Autoencoders for end-to-end mapping of orthophotos to road segmentations. In addition, a set of post-processing steps were applied to make the model outputs GIS-ready data that could be useful for various applications. The optimization of the model`s parameters is explained which was conducted via grid search method. The model was trained and implemented in Keras, a high-level deep learning framework run on top of Tensorflow. The results show that the proposed model with the best-obtained hyperparameters could segment road objects from orthophotos at an average accuracy of 88.5%. The results of optimization revealed that the best optimization algorithm and activation function for the studied task are Stochastic Gradient Descent (SGD) and Exponential Linear Unit (ELU), respectively. In addition, the best numbers of convolutional filters were found to be 8 for the first and second layers and 128 for the third and fourth layers of the proposed network architecture. Moreover, the analysis on the time complexity of the model showed that the model could be trained in 4 hours and 50 minutes on 1024 high-resolution images of size 106 × 106 pixels, and segment road objects from similar size and resolution images in around 14 minutes. The results show that the deep learning models such as Convolutional Autoencoders could be a best alternative to traditional machine learning models for road segmentation from aerial photographs.

A Novel Road Segmentation Technique from Orthophotos Using Deep Convolutional Autoencoders

Sameen, Maher Ibrahim,Pradhan, Biswajeet The Korean Society of Remote Sensing 2017 大韓遠隔探査學會誌 Vol.33 No.4

This paper presents a deep learning-based road segmentation framework from very high-resolution orthophotos. The proposed method uses Deep Convolutional Autoencoders for end-to-end mapping of orthophotos to road segmentations. In addition, a set of post-processing steps were applied to make the model outputs GIS-ready data that could be useful for various applications. The optimization of the model's parameters is explained which was conducted via grid search method. The model was trained and implemented in Keras, a high-level deep learning framework run on top of Tensorflow. The results show that the proposed model with the best-obtained hyperparameters could segment road objects from orthophotos at an average accuracy of 88.5%. The results of optimization revealed that the best optimization algorithm and activation function for the studied task are Stochastic Gradient Descent (SGD) and Exponential Linear Unit (ELU), respectively. In addition, the best numbers of convolutional filters were found to be 8 for the first and second layers and 128 for the third and fourth layers of the proposed network architecture. Moreover, the analysis on the time complexity of the model showed that the model could be trained in 4 hours and 50 minutes on 1024 high-resolution images of size $106{\times}106pixels$, and segment road objects from similar size and resolution images in around 14 minutes. The results show that the deep learning models such as Convolutional Autoencoders could be a best alternative to traditional machine learning models for road segmentation from aerial photographs.