• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.41 no.10
• /
• pp.931-940
• /
• 2017

This paper describes a predictive fault diagnosis algorithm for autonomous vehicles based on a kinematic model that uses a sliding mode observer. To ensure the safety of autonomous vehicles, reliable information about the environment and vehicle dynamic states is required. A predictive algorithm that can interactively diagnose longitudinal environment and vehicle acceleration information is proposed in this paper to evaluate the reliability of sensors. To design the diagnosis algorithm, a longitudinal kinematic model is used based on a sliding mode observer. The reliability of the fault diagnosis algorithm can be ensured because the sliding mode observer utilized can reconstruct the relative acceleration despite faulty signals in the longitudinal environment information. Actual data based performance evaluations are conducted with various fault conditions for a reasonable performance evaluation of the predictive fault diagnosis algorithm presented in this paper. The evaluation results show that the proposed diagnosis algorithm can reasonably diagnose the faults in the longitudinal environment and acceleration information for all fault conditions.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.4
• /
• pp.28-46
• /
• 2021

85-90% of the causes of traffic accidents are human factors, and autonomous vehicles with little free running distance can be an alternative to prevent traffic accidents caused by human factors. However, securing safety of autonomous vehicles should be preceded in order to reduce traffic accident damage through the introduction of autonomous vehicles. Therefore, it is necessary to verify whether the vehicle can respond appropriately to changes in the road and traffic environment through repeated and reproduced test runs in an environment similar to the actual road. In this study, K-City's development strategies for upgrading, differentiating, and systematic development were established by comparing and analyzing the current status of domestic and foreign testbeds and business environment analysis. Furthermore, we derive challenge tasks to achieve each strategy.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.6
• /
• pp.162-176
• /
• 2021

Recently, Autonomous Vehicle(AV) has been expected to solve various transportation problems. s the problem of environmental pollution become serious, research to reduce pollution is needed. However, empirical research on AV related pollution is insufficient. Based on this background, this study analyzed network performance changes and CO₂ emissions introduc AVs and Electric Vehicles(EV) in eight intersections. The results show that when AVs with internal combustion engines were, the effect of carbon reduction over the network was insignificant. On the other hand, it was that the total amount of CO₂ generated in the network decreased significantly when EVs and autonomous electric vehicles were emissions in the transportation sector.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.75-81
• /
• 2022

Autonomous vehicles, which can dramatically solve the lack of parking spaces, are making great progress through deep reinforcement learning. Activation functions are used for deep reinforcement learning, and various activation functions have been proposed, but their performance deviations were large depending on the application environment. Therefore, finding the optimal activation function depending on the environment is important for effective learning. This paper analyzes 12 functions mainly used in reinforcement learning to compare and evaluate which activation function is most effective when autonomous vehicles use deep reinforcement learning to learn parking. To this end, a performance evaluation environment was established, and the average reward of each activation function was compared with the success rate, episode length, and vehicle speed. As a result, the highest reward was the case of using GELU, and the ELU was the lowest. The reward difference between the two activation functions was 35.2%.

• Journal of Drive and Control
• /
• v.19 no.1
• /
• pp.77-80
• /
• 2022

• Journal of the Korea Convergence Society
• /
• v.11 no.10
• /
• pp.71-79
• /
• 2020

Due to the recent changes in residential value, user needs for various spaces in residential space are increasing. However, it is not easy to change and reconstruct the first residential form according to the needs of the user due to the characteristics of the residential space where movement and deformation are difficult. Therefore, this study aims to derive a plan to meet user needs according to changing residential value by using fully autonomous vehicles with a concept of space that can be moved. First, the existing residential space was classified through literature research, and based on this, the space type of the fully autonomous vehicle for the residential was derived through user research. Then, through the survey method and statistical analysis, user needs analysis for residential space, usefulness analysis of derived space type and type specification were carried out. Finally, based on the results of the study, the design direction of the fully autonomous vehicle for residential purpose was presented. Based on the user's perception, it is meaningful that it categorizes the space of the fully autonomous vehicle for residential purposes and suggests the design direction.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.1
• /
• pp.205-213
• /
• 2017

This paper is a study on the method to derive the functional level required for autonomous unmanned ground vehicle, one of the defense robots. Conventional weapon systems are not significantly affected by the operating environment, while defense robots exhibit different performance depending on the operating environment, even if they are on the same platform. If the performance of defense robot is different depending on operational environment, results of mission performance will be vary significantly. Therefore, it is necessary to clarify the level of function required by the military in order to research and develop most optimal defense robots. In this thesis, we propose a method to derive the required function level of unmanned ground vehicles, focusing on autonomous driving, one of the most vital functions of defense robots. Our results showed that the autonomous driving function depending intervention levels and evaluated functional sensitivity for autonomous driving of the unmanned vehicle using climate and topography as variables.

• Journal of the Korea Convergence Society
• /
• v.10 no.11
• /
• pp.95-100
• /
• 2019

In this paper, autonomous vehicles of service with existing vehicle accident for the prevention of the vehicle communication technology, self-driving techniques, brakes automatic control technology, artificial intelligence technologies such as well and developed the vehicle accident this occur to death or has been techniques, can prepare various safety cases intended to minimize the injury. In this paper, it is a study to secure safety in autonomous vehicles. This is determined according to spatial factors such as chip signals for general low-power short-range wireless communication and micro road AI. On the other hand, in this paper, the safety of boarding is improved by checking the signal from the electronic chip, up to "recognition of the emotion from residence time in the sensing area" to the biological electronic chip. As a result of demonstrating the reliability of the world countries the world, inducing safety autonomous system of all passengers in terms of safety. Unmanned autonomous vehicle riding and commercialization will lead to AI systems and biochips (Verification), linked IoT on the road in the near future, and the safety technology reliability of the world will be highlighted.

• International conference on construction engineering and project management
• /
• 2020.12a
• /
• pp.510-519
• /
• 2020

Within an intelligent automated cyber-physical system, the realization of the autonomous mechanism for data collection, data integration, and data analysis plays a critical role in the design, development, operation, and maintenance of such a system. This construct is particularly vital for fault-tolerant route-finding systems that rely on the imprecise GPS location of the vehicles to properly operate, timely plan, and continuously produce informative feedback to the user. More essentially, the integration of digital twins with cyber-physical route-finding systems has been overlooked in intelligent transportation services with the capacity to construct the network routes solely from the locations of the operating vehicles. To address this limitation, the present study proposes a conceptual architecture that employs digital twin to autonomously maintain, update, and manage intelligent transportation systems. This virtual management simulation can improve the accuracy of time-of-arrival prediction based on auto-generated routes on which the vehicle's real-time location is mapped. To that end, first, an intelligent transportation system was developed based on two primary mechanisms: 1) an automated route finding process in which predictive data-driven models (i.e., regularized least-squares regression) can elicit the geometry and direction of the routes of the transportation network from the cloud of geotagged data points of the operating vehicles and 2) an intelligent mapping process capable of accurately locating the vehicles on the map whereby their arrival times to any point on the route can be estimated. Afterward, the digital representations of the physical entities (i.e., vehicles and routes) were simulated based on the auto-generated routes and the vehicles' locations in near-real-time. Finally, the feasibility and usability of the presented conceptual framework were evaluated through the comparison between the primary characteristics of the physical entities with their digital representations. The proposed architecture can be used by the vehicle-tracking applications dependent on geotagged data for digital mapping and location tracking of vehicles under a systematic comparison and simulation cyber-physical system.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.24 no.2
• /
• pp.102-116
• /
• 2021

For the safe operation of autonomous vehicles, the representative technology of the 4^th industrial revolution era, a combination of various technologies such as sensor technology, software technology and car technology is required. An autonomous vehicle is a vehicle that recognizes current location and situation by using the various sensors, and makes its own decisions without depending on the driver. Perfect recognition technology is required for fully autonomous driving. Since the precise road maps provide various road information including lanes, stop lines, traffic lights and crosswalks, it is possible to minimize the cognitive errors that occur in autonomous vehicles by using the precise road maps with location information of the road facilities. In this study, the definition, necessity and technical trends of the precise road map have been analyzed, and the HD(High Definition) map prototype based on the web GIS has been built in the autonomous driving-specialized areas of Daegu Metropolitan City(Suseong Medical District, about 24km), the Happy City of Sejong Special Self-Governing City(about 33km), and the FMTC(Future Mobility Technical Center) PG(Proving Ground) of Seoul National University Siheung Campus using the MMS(Mobile Mapping System) surveying results given by the National Geographic Information Institute. In future research, the built-in precise road map service will be installed in the autonomous vehicles and control systems to verify the real-time locations and its location correction algorithm.