RVCar: An FPGA-Based Simple and Open-Source Mini Motor Car System with a RISC-V Soft Processor

Takuto KANAMORI; Takashi ODAN; Kazuki HIROHATA; Kenji KISE

doi:10.1587/transinf.2022PAP0004

IEICE TRANSACTIONS on Information

RVCar: An FPGA-Based Simple and Open-Source Mini Motor Car System with a RISC-V Soft Processor

Takuto KANAMORI, Takashi ODAN, Kazuki HIROHATA, Kenji KISE

Full Text Views

0

Cite this

Summary :

Deep Neural Network (DNN) is widely used for computer vision tasks, such as image classification, object detection, and segmentation. DNN accelerator on FPGA and especially Convolutional Neural Network (CNN) is a hot topic. More research and education should be conducted to boost this field. A starting point is required to make it easy for new entrants to join this field. We believe that FPGA-based Autonomous Driving (AD) motor cars are suitable for this because DNN accelerators can be used for image processing with low latency. In this paper, we propose an FPGA-based simple and open-source mini motor car system named RVCar with a RISC-V soft processor and a CNN accelerator. RVCar is suitable for the new entrants who want to learn the implementation of a CNN accelerator and the surrounding system. The motor car consists of Xilinx Nexys A7 board and simple parts. All modules except the CNN accelerator are implemented in Verilog HDL and SystemVerilog. The CNN accelerator is converted from a PyTorch model by our tool. The accelerator is written in C++, synthesizable by Vitis HLS, and an easy-to-customize baseline for the new entrants. FreeRTOS is used to implement AD algorithms and executed on the RISC-V soft processor. It helps the users to develop the AD algorithms efficiently. We conduct a case study of the simple AD task we define. Although the task is simple, it is difficult to achieve without image recognition. We confirm that RVCar can recognize objects and make correct decisions based on the results.

Publication: IEICE TRANSACTIONS on Information Vol.E105-D No.12 pp.1999-2007

Publication Date: 2022/12/01

Publicized: 2022/08/09

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2022PAP0004

Type of Manuscript: Special Section PAPER (Special Section on Forefront Computing)

Category

Authors

Takuto KANAMORI
  Tokyo Institute of Technology
Takashi ODAN
  Tokyo Institute of Technology
Kazuki HIROHATA
  Tokyo Institute of Technology
Kenji KISE
  Tokyo Institute of Technology

Keyword

FPGA, motor car, Nexys A7, CNN accelerator, RISC-V, soft processor, real time operating system, freeRTOS

Cite this

Copy

Takuto KANAMORI, Takashi ODAN, Kazuki HIROHATA, Kenji KISE, "RVCar: An FPGA-Based Simple and Open-Source Mini Motor Car System with a RISC-V Soft Processor" in IEICE TRANSACTIONS on Information, vol. E105-D, no. 12, pp. 1999-2007, December 2022, doi: 10.1587/transinf.2022PAP0004.
Abstract: Deep Neural Network (DNN) is widely used for computer vision tasks, such as image classification, object detection, and segmentation. DNN accelerator on FPGA and especially Convolutional Neural Network (CNN) is a hot topic. More research and education should be conducted to boost this field. A starting point is required to make it easy for new entrants to join this field. We believe that FPGA-based Autonomous Driving (AD) motor cars are suitable for this because DNN accelerators can be used for image processing with low latency. In this paper, we propose an FPGA-based simple and open-source mini motor car system named RVCar with a RISC-V soft processor and a CNN accelerator. RVCar is suitable for the new entrants who want to learn the implementation of a CNN accelerator and the surrounding system. The motor car consists of Xilinx Nexys A7 board and simple parts. All modules except the CNN accelerator are implemented in Verilog HDL and SystemVerilog. The CNN accelerator is converted from a PyTorch model by our tool. The accelerator is written in C++, synthesizable by Vitis HLS, and an easy-to-customize baseline for the new entrants. FreeRTOS is used to implement AD algorithms and executed on the RISC-V soft processor. It helps the users to develop the AD algorithms efficiently. We conduct a case study of the simple AD task we define. Although the task is simple, it is difficult to achieve without image recognition. We confirm that RVCar can recognize objects and make correct decisions based on the results.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022PAP0004/_p

Copy

@ARTICLE{e105-d_12_1999,
author={Takuto KANAMORI, Takashi ODAN, Kazuki HIROHATA, Kenji KISE, },
journal={IEICE TRANSACTIONS on Information},
title={RVCar: An FPGA-Based Simple and Open-Source Mini Motor Car System with a RISC-V Soft Processor},
year={2022},
volume={E105-D},
number={12},
pages={1999-2007},
abstract={Deep Neural Network (DNN) is widely used for computer vision tasks, such as image classification, object detection, and segmentation. DNN accelerator on FPGA and especially Convolutional Neural Network (CNN) is a hot topic. More research and education should be conducted to boost this field. A starting point is required to make it easy for new entrants to join this field. We believe that FPGA-based Autonomous Driving (AD) motor cars are suitable for this because DNN accelerators can be used for image processing with low latency. In this paper, we propose an FPGA-based simple and open-source mini motor car system named RVCar with a RISC-V soft processor and a CNN accelerator. RVCar is suitable for the new entrants who want to learn the implementation of a CNN accelerator and the surrounding system. The motor car consists of Xilinx Nexys A7 board and simple parts. All modules except the CNN accelerator are implemented in Verilog HDL and SystemVerilog. The CNN accelerator is converted from a PyTorch model by our tool. The accelerator is written in C++, synthesizable by Vitis HLS, and an easy-to-customize baseline for the new entrants. FreeRTOS is used to implement AD algorithms and executed on the RISC-V soft processor. It helps the users to develop the AD algorithms efficiently. We conduct a case study of the simple AD task we define. Although the task is simple, it is difficult to achieve without image recognition. We confirm that RVCar can recognize objects and make correct decisions based on the results.},
keywords={},
doi={10.1587/transinf.2022PAP0004},
ISSN={1745-1361},
month={December},}

Copy

TY - JOUR
TI - RVCar: An FPGA-Based Simple and Open-Source Mini Motor Car System with a RISC-V Soft Processor
T2 - IEICE TRANSACTIONS on Information
SP - 1999
EP - 2007
AU - Takuto KANAMORI
AU - Takashi ODAN
AU - Kazuki HIROHATA
AU - Kenji KISE
PY - 2022
DO - 10.1587/transinf.2022PAP0004
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E105-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2022
AB - Deep Neural Network (DNN) is widely used for computer vision tasks, such as image classification, object detection, and segmentation. DNN accelerator on FPGA and especially Convolutional Neural Network (CNN) is a hot topic. More research and education should be conducted to boost this field. A starting point is required to make it easy for new entrants to join this field. We believe that FPGA-based Autonomous Driving (AD) motor cars are suitable for this because DNN accelerators can be used for image processing with low latency. In this paper, we propose an FPGA-based simple and open-source mini motor car system named RVCar with a RISC-V soft processor and a CNN accelerator. RVCar is suitable for the new entrants who want to learn the implementation of a CNN accelerator and the surrounding system. The motor car consists of Xilinx Nexys A7 board and simple parts. All modules except the CNN accelerator are implemented in Verilog HDL and SystemVerilog. The CNN accelerator is converted from a PyTorch model by our tool. The accelerator is written in C++, synthesizable by Vitis HLS, and an easy-to-customize baseline for the new entrants. FreeRTOS is used to implement AD algorithms and executed on the RISC-V soft processor. It helps the users to develop the AD algorithms efficiently. We conduct a case study of the simple AD task we define. Although the task is simple, it is difficult to achieve without image recognition. We confirm that RVCar can recognize objects and make correct decisions based on the results.
ER -

IEICE TRANSACTIONS on Information