• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.970-977
• /
• 2018

The autonomous vehicle is based on an advanced driver assistance system (ADAS) consisting of a sensor that collects information about the surrounding environment and a control module that determines the measured data. As interest in autonomous navigation technology grows recently, an easy development framework for ADAS beginners and learners is needed. However, existing development and verification methods are based on high performance vehicle simulator, which has drawbacks such as complexity of verification method and high cost. Also, most of the schemes do not provide the sensing data required by the ADAS directly from the simulator, which limits verification reliability. In this paper, we present an interactive ADAS development and verification framework using a 3D vehicle simulator that overcomes the problems of existing methods. ADAS with image recognition based artificial intelligence was implemented as a virtual sensor in a 3D car simulator, and autonomous driving verification was performed in real scenarios.

• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.1073-1078
• /
• 2018

As interest in the 4th Industrial Revolution increases, the CPS, in which things existing in the reality and things existing in the virtual interact with each other, is attracting attention as an important technology. Complex systems such as electric vehicles, autonomous driving, smart factories, and smart grid system are considered as core technology fields of the 4th Industrial Revolution, and many types of research have been conducted to develop it. The reliability of the system is directly related to the safety of people in case of the autonomous driving, and verification of the actual vehicle's hardware and software of ADAS is essential. In this paper, we proposed distributed simulation middleware supporting HILS for reliable verification of the complex hybrid systems.

• Journal of IKEEE
• /
• v.20 no.3
• /
• pp.241-250
• /
• 2016

AVM(Around View Monitoring) is a function of ADAS(Advanced Driver Assistance Systems), which provides a bird's eye view of the surroundings of a vehicle to the user. AVM systems require large bandwidth since they are composed of four input images and require real-time processing for vehicle-embedded environments. Also, the memory bandwidth requirement increases greatly when the resolution of the input data is higher. In this paper, we propose four basic hardware models of AVM systems. The models are decided by whether or not there is a valid data extraction module and an image processing purpose LUT generation module. We analyze the required bandwidth and hardware resource for each model. For verification of the proposed models, we implemented an AVM system using XC7Z045 FPGA and DDR3 memory for VGA and FHD resolution. All four of the proposed hardware model is executed below 33ms, which shows that it can operate in real-time.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.3
• /
• pp.57-64
• /
• 2020

The leading technologies of the ADAS (Advanced Driver Assist System) are ACC (Advanced Cruise Control), LKAS (Lane Keeping Assist System), and AEB (Autonomous Emergency Braking). LKAS is a system that uses cameras and infrared sensors to control steering and return to its running lane in the event of unintentional deviations. The actual test is performed for a safety evaluation and verification of the system. On the other hand, research on the system evaluation method is insufficient when an additional steering angle is applied. In this study, a model using Prescan was developed and simulated for the scenarios proposed in the preceding study. Comparative analyses of the simulation and the actual test were performed. As a result, the modeling validity was verified. A difference between the front wheels and the lane occurred due to the return velocity. The results revealed a maximum error of 0.56 m. The error occurred because the lateral velocity of the car was relatively small. On the other hand, the distance from wheels to the lanes displayed a tendency of approximately 0.5 m. This can be verified reliably.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.22 no.5
• /
• pp.126-135
• /
• 2014

This paper describes the development and verification of control algorithms for V2X and environmental sensor integrated intersection support and safety systems. The objective of the research is to develop core technologies for effective fusion of V2X and environmental sensors and to develop new safety function for intersection safety. One of core technologies is to achieve the improvement of GPS accuracy, and the other is to develop the algorithm of a vehicle identification which matches all data from V2X, vehicle sensors and environmental sensors to specific vehicles. A intersection optimal pass (IOP) algorithm is designed based on these core technologies. IOP recommends appropriate speed to pass the intersection in the consideration of traffic light signal and preceeding vehicle existence. Another function is developed to prevent a collision avoidance when car crash caused by traffic violation of surrounding vehicles is expected. Finally all functions are implemented and tested in three test vehicles. It is shown that IOP can support convenient and comfortable driving with recommending optimal pass speed and collision avoidance algorithm can effectively prevent collision caused by traffic sign violation of surrounding vehicles.

• Journal of Advanced Navigation Technology
• /
• v.19 no.6
• /
• pp.544-551
• /
• 2015

In this paper, the design and implementation results of the optical flow estimator (OFE) for moving object detection (MOD) in advanced driver assistance system (ADAS). In the proposed design, Brox's algorithm with global optimization is considered, which shows the high performance in the vehicle environment. In addition, Cholesky factorization is applied to solve Euler-Lagrange equation in Brox's algorithm. Also, shift register bank is incorporated to reduce memory access rate. The proposed optical flow estimator was designed with Verilog-HDL, and FPGA board was used for the real-time verification. Implementation results show that the proposed optical flow estimator includes the logic slices of 40.4K, 155 DSP48s, and block memory of 11,290Kbits.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.1
• /
• pp.126-136
• /
• 2014

In the Advanced Driver Assistance Systems (ADAS) of smart vehicle and Intelligent Transportation System (ITS) for to detect the boundary of lane is being studied a lot of Hough Transform. This method detects correctly recognition the lane. But recognition rate can fall due to detecting straight lines outside of the lane. In order to solve this problems, this paper proposed an algorithm to recognize the lane boundaries and the accumulator cells in Hough space. Based on proposed algorithm, we develop application for Android was developed by H/W verification. Users of smart phone devices could use lane detection and lane departure warning systems for driver's safety whenever and wherever. Software verification using the OpenCV showed efficiency recognition correct rate of 93.8% and hardware real-time verification for an application development in the Android phone showed recognition correct rate of 70%.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.06a
• /
• pp.267-270
• /
• 2016

차량 인식 기술은 지능형 자율주행 차량 및 첨단 운전자 보조 시스템 (ADAS: Advanced Driver Assistance System)의 개발에 있어서 핵심 요소 기술이다. 영상 기반의 차량 검출 알고리즘은 일반적으로 가설 생성 (HG: Hypothesis Generation) 단계와 가설 검증 (HV: Hypothesis Verification) 단계로 구성된다. 가설 검증 단계는 관심 영역 (ROI: Region of Interest) 내에 차량이 존재할 가능성이 있는 후보 영역을 만드는 단계로서 전체 알고리즘의 복잡도와 성능에 영향을 미친다. 본 논문에서는 관심 영역 내에 존재하는 그림자와 차량으로 인한 에지를 검출하고 두 특징 정보를 결합한 가설 생성 방법을 제안하고 차량 후방 영상을 이용하여 사각지대를 감시하는 시스템에 제안 방법을 적용하는 실험을 수행하였다. 실험 결과로 제안 방법이 차량 후보 영역의 존재 여부와 위치 정보를 판단하기에 적합하며 이를 통해 차량 검출 알고리즘의 계산 복잡도를 개선하면서도 다음 단계인 가설 검증 시 검출 성능을 향상시킬 수 있음을 확인하였다.