• Title/Summary/Keyword: Hardware acceleration

Search Result 114, Processing Time 0.019 seconds

Development of engine control based TCS slip control algorithm using engine map (엔진맵에 기초한 엔진제어 TCS 슬립제어 알고리듬의 개발)

  • Song, Jae-Bok;Kim, Byeong-Cheol
    • Transactions of the Korean Society of Mechanical Engineers A
    • /
    • v.22 no.2
    • /
    • pp.428-436
    • /
    • 1998
  • A TCS slip control system improves acceleration capability and steerability on slippery roads through engine torgue and/or brake torque control. This research mainly deals with the engine control algorithm via the adjustment of the engine throttle angle. The following new control strategy is proposed and investigated ; the TCS slip controller whose input is the difference between the desired driving wheel speed corresponding to the optimum slip ratio and the actual speed yields the target engine torque and then estimates the throttle angle based on the engine performance curve. Various simulation and hardware-in-the-loop simulation have been carried out. The results show the proposed strategy may compensate for the inherent nonlinearity between variation of the throttle angle and variation of the engine torque and produce better performance than the previous strategies without the engine map, especially in the high speed region.

Development of HILS System for Performance Analysis of the ABS ECU for Commercial Vehicles (상용차용 ABS ECU의 성능분석을 위한 HILS 시스템 개발)

  • 황돈하;이기창;전정우;김용주;조정목;조중선
    • Journal of Institute of Control, Robotics and Systems
    • /
    • v.8 no.10
    • /
    • pp.898-906
    • /
    • 2002
  • Antilock Brake System (ABS) is designed to prevent wheels from being locked-up under emergency braking of a vehicle. Therefore it improves directional stability of the vehicle, shortens stopping distance, and enhances maneuvering during braking, regardless of road conditions. Hardware In-the-Loop Simulation (HILS) is an effective tool for design Performance evaluation and test of vehicle subsystems such as ABS, active suspension, and steering systems. This paper describes a HILS model for ABS/ ASR(Acceleration Slip Regulation) system applications. A fourteen degrees-of-freedom vehicle dynamics model is simulated in an alpha-chip processor board. The proposed HILS system is tested with a basic ABS control algorithm. The design and implementation of HILS system for the ABS ECU(Electronic Control Unit) development of commercial vehicle are presented. The results show that the proposed HILS system can be used to test the performance, stability, and reliability of a vehicle under braking.

Performance Evaluation of a Mixed-Mode Type ER Engine Mount(II)-Performance Evaluation Via HILS- (복합 모드형 ER엔진마운트의 성능평가 (II) - HILL를 통한 성능 평가 -)

  • Choe, Yeong-Tae;Choe, Seung-Bok
    • Transactions of the Korean Society of Mechanical Engineers A
    • /
    • v.24 no.9 s.180
    • /
    • pp.2151-2158
    • /
    • 2000
  • This paper presents vibration control performance of a passenger vehicle installed with the mixed-mode type ER engine mounts. The performance is evaluated via hardware-in-the-loop-simulation(HILS) method. As a first step, a dynamic model of a vehicle featuring the ER engine mounts is formulated by taking into account the engine excitation forces. A new type of the fuzzy skyhook controller is then established in order to control both engine and body vibrations. This is accomplished by adopting a weighting parameter between two performance criteria which is to be determined from the fuzzy algorithm. Vertical displacement and acceleration of the engine mount obtained from the HILS method are provided in the frequency domain. In addition, vibration control performance between the conventional hydraulic engine mount and the proposed engine mount is compared in the time and frequency domains.

A View-Frustum Culling Technique Using OpenGL for Large Polygon Models (OpenGL을 이용한 대용량 Polygon Model의 View-Frustum Culling 기법)

  • Cho, Doo-Yeoun;Jung, Sung-Jun;Lee, Kyu-Yeul;Kim, Tae-Wan;Choi, Hang-Soon;Seong, Woo-Jae
    • Journal of Korea Game Society
    • /
    • v.1 no.1
    • /
    • pp.55-60
    • /
    • 2001
  • With rapid development of graphic hardware, researches on Virtual Reality and 3D Games have received more attention than before. For more realistic 3D graphic scene, objects were to be presented with lots of polygons and the number of objects shown in a scene was remarkably increased. Therefore, for effective visualization of large polygon models like this, view-frustum culling method, that visualizes only objects shown in the screen, has been widely used. In general, the bounding boxes that include objects are generated firstly, and the boxes are intersected with view-frustum to check whether object is in the visible area or not. Recently, an algorithm that can check in-out test of objects using OpenGL's selection mode, which is originally used to select the objects in the screen, is suggested. This algorithm is fast because it can use hardware acceleration. In this study, by implementing and applying this algorithm to large polygon models, we showed the efficiency of OpenGL assisted View-Frustum Culling algorithm. If this algorithm is applied to 3D games that have to process more complicated characters and landscapes, performance improvement can be expected.

  • PDF

Development of Navigation HILS System for Integrated Navigation Performance Analysis of Large Diameter Unmanned Underwater Vehicle (LDUUV) (대형급 탐색용 무인잠수정 복합항법 성능 분석을 위한 항법 HILS 시스템 개발)

  • Yoo, Tae-Suk;Kim, Moon Hwan;Hwang, Jong Hyun;Yoon, Seon Il
    • Journal of Ocean Engineering and Technology
    • /
    • v.30 no.5
    • /
    • pp.367-373
    • /
    • 2016
  • This paper describes the development of a navigation HILS (hardware in the loop simulation) system for an integrated navigation performance analysis of a large diameter unmanned underwater vehicle (LDUUV). The HILS system was used for the performance analysis of the LDUUV. When a conventional HILS system is used, it is not possible to calculate the velocity and position using an inertial navigation system (INS). To cope with this problem, an external acceleration was generated. To evaluate the proposed method, we compare the results of a Monte Carlo simulation and navigation HILS experiment.

Design and Implementation of Multi-mode Sensor Signal Processor on FPGA Device (다중모드 센서 신호 처리 프로세서의 FPGA 기반 설계 및 구현)

  • Soongyu Kang;Yunho Jung
    • Journal of Sensor Science and Technology
    • /
    • v.32 no.4
    • /
    • pp.246-251
    • /
    • 2023
  • Internet of Things (IoT) systems process signals from various sensors using signal processing algorithms suitable for the signal characteristics. To analyze complex signals, these systems usually use signal processing algorithms in the frequency domain, such as fast Fourier transform (FFT), filtering, and short-time Fourier transform (STFT). In this study, we propose a multi-mode sensor signal processor (SSP) accelerator with an FFT-based hardware design. The FFT processor in the proposed SSP is designed with a radix-2 single-path delay feedback (R2SDF) pipeline architecture for high-speed operation. Moreover, based on this FFT processor, the proposed SSP can perform filtering and STFT operation. The proposed SSP is implemented on a field-programmable gate array (FPGA). By sharing the FFT processor for each algorithm, the required hardware resources are significantly reduced. The proposed SSP is implemented and verified on Xilinxh's Zynq Ultrascale+ MPSoC ZCU104 with 53,591 look-up tables (LUTs), 71,451 flip-flops (FFs), and 44 digital signal processors (DSPs). The FFT, filtering, and STFT algorithm implementations on the proposed SSP achieve 185x average acceleration.

Structured Pruning for Efficient Transformer Model compression (효율적인 Transformer 모델 경량화를 위한 구조화된 프루닝)

  • Eunji Yoo;Youngjoo Lee
    • Transactions on Semiconductor Engineering
    • /
    • v.1 no.1
    • /
    • pp.23-30
    • /
    • 2023
  • With the recent development of Generative AI technology by IT giants, the size of the transformer model is increasing exponentially over trillion won. In order to continuously enable these AI services, it is essential to reduce the weight of the model. In this paper, we find a hardware-friendly structured pruning pattern and propose a lightweight method of the transformer model. Since compression proceeds by utilizing the characteristics of the model algorithm, the size of the model can be reduced and performance can be maintained as much as possible. Experiments show that the structured pruning proposed when pruning GPT-2 and BERT language models shows almost similar performance to fine-grained pruning even in highly sparse regions. This approach reduces model parameters by 80% and allows hardware acceleration in structured form with 0.003% accuracy loss compared to fine-tuned pruning.

AB9: A neural processor for inference acceleration

  • Cho, Yong Cheol Peter;Chung, Jaehoon;Yang, Jeongmin;Lyuh, Chun-Gi;Kim, HyunMi;Kim, Chan;Ham, Je-seok;Choi, Minseok;Shin, Kyoungseon;Han, Jinho;Kwon, Youngsu
    • ETRI Journal
    • /
    • v.42 no.4
    • /
    • pp.491-504
    • /
    • 2020
  • We present AB9, a neural processor for inference acceleration. AB9 consists of a systolic tensor core (STC) neural network accelerator designed to accelerate artificial intelligence applications by exploiting the data reuse and parallelism characteristics inherent in neural networks while providing fast access to large on-chip memory. Complementing the hardware is an intuitive and user-friendly development environment that includes a simulator and an implementation flow that provides a high degree of programmability with a short development time. Along with a 40-TFLOP STC that includes 32k arithmetic units and over 36 MB of on-chip SRAM, our baseline implementation of AB9 consists of a 1-GHz quad-core setup with other various industry-standard peripheral intellectual properties. The acceleration performance and power efficiency were evaluated using YOLOv2, and the results show that AB9 has superior performance and power efficiency to that of a general-purpose graphics processing unit implementation. AB9 has been taped out in the TSMC 28-nm process with a chip size of 17 × 23 ㎟. Delivery is expected later this year.

Development of Fast Side-impact Sensing Algorithm (고속 측면 충돌 감지 알고리즘의 개발)

  • 박서욱;김현태
    • Transactions of the Korean Society of Automotive Engineers
    • /
    • v.8 no.3
    • /
    • pp.163-170
    • /
    • 2000
  • Accident statistics shows that the portion of fatal occupant injuries due to side impacts is considerably high. The side impact usually leads to a severe intrusion of side structure into the passenger compartment. Furthermore, the safety zone for the side impact is relatively small compared to the front impact. Those kinds of physics for side impact frequently result in a fatal injury for the occupant. Therefore, NHTSA and EEVC are trying to intensify the regulation for the occupant protection against side impact. Both the regulation and recent market trends are asking for an installation of side airbag. There are several types of system configuration for side impact sensing. In this paper, we adopt the acceleration-based remote sensing method for the side airbag control system. We mainly focus on the development of hardware and crash discrimination algorithm of remote sensing unit. The crash discrimination algorithm needs fast decision of airbag firing especially for high-speed side impact such as FMVSS 214 and EEVC tests. It is also required to distinguish between low-speed fire and no-fire events. The algorithm should have a sufficient safety margin against any misuse situation such as hammer blow, door slam, etc. This paper introduces several firing criteria such as acceleration. velocity and energy criteria that use physical value proportional to crash severity. We have made a simulation program by using Matlab/Simulink to implement the proposed algorithm. We have conducted an algorithm calibration by using real crash data for 2,500cc vehicle. The crash performance obtained by the simulation was verified through a pulse injection method. It turned out that the results satisfied the system requirements well.

  • PDF

A Study on the Appication of Semi-Active Supension Units for a Combat Vehicle by Using HILS (HILS를 활용한 전투차량의 반능동 현수장치 적용에 관한 연구)

  • Kim, Chi-Ung;Kim, Moon-June;Rhee, Eun-Jun;Lee, Kyoung-Hoon;Woo, Kwan-Je
    • Journal of the Korea Institute of Military Science and Technology
    • /
    • v.13 no.6
    • /
    • pp.967-975
    • /
    • 2010
  • There have been a lot of efforts on the improvement for the ride comfort and handling stability of the combat vehicles. Especially most of vehicles for military purpose have bad inertial condition and severe operating condition such as the rough road driving, and need a high mobility in the emergency status. It is necessary to apply the controlled suspension system in order to improve the vehicle mobile stability and ride comfort ability of crews. A feasibility study is performed on the application of the semi-active suspension system with a magneto-rheological controlled shock absorber for a $6{\times}6$ combat vehicle. First, the dynamic simulation model of the vehicle including the control model for the semi-active suspension system was executed. Based on this model, a hardware-in-the-loop simulation(HILS) system which has a semi-active suspension controller hardware was constructed. After full vehicle simulations were performed in virtual proving courses with this system, the semi-active suspension system was proven to give better ride comfort and handling stability in comparison with the conventional passive suspension system.