Low-Power Magnetic Displacement Sensor Based on RISC-V Embedded System

Tao Sun, Yue Song, Huiyun Yang2024 · DOI

TL;DR

This paper presents a low-power magnetic displacement sensor system built on the RISC-V embedded architecture. By porting Hall sensor technology to RISC-V, the researchers demonstrate a system that achieves comparable performance to traditional x86 platforms while significantly reducing power consumption, making it ideal for industrial applications.

What Problem Does This Paper Address?

The increasing demand for energy-efficient and cost-effective sensing solutions in industrial settings presents a challenge. While RISC-V architecture offers inherent low-power advantages and flexibility, its integration with magnetic displacement sensing, specifically using Hall sensors, has not been extensively explored. Traditional systems often rely on more power-hungry architectures. This paper aims to bridge this gap by demonstrating the feasibility and effectiveness of combining Hall sensors with RISC-V for displacement sensing, promising a more sustainable and adaptable solution.

Methods

The core of this research involves porting Hall sensor functionality to a RISC-V embedded system. This includes:

  • RISC-V Development Environment: Utilizing the RISC-V GNU Compiler Toolchain and Software Development Kits (SDKs) to create an embedded development environment. This allows for cross-compilation and interaction with hardware peripherals.
  • Hall Sensor Driver Porting: Adapting existing Hall sensor driver code to the RISC-V architecture. This involves understanding the sensor's communication protocol (analog output in this case, connected to an ADC) and implementing the necessary firmware to read and process its signals.
  • PM-MFM Method Implementation: The study focuses on the Permanent Magnet Magnetic Flux Measurement (PM-MFM) method. This technique uses permanent magnets to create a magnetic field, and a Hall sensor measures the changes in magnetic flux density caused by the movement of a ferromagnetic target. The firmware implements algorithms to convert these magnetic field measurements into displacement values.
  • Hardware Interface Design: Careful consideration is given to the interface between the RISC-V microcontroller and the Hall sensor module, focusing on signal integrity, noise immunity, and stable power supply.

Datasets

Not discussed in the paper. The experiments utilize a self-constructed experimental setup with a steel plate as the target object.

Key Insights and Ideas

  • RISC-V for Low-Power Sensing: The research strongly validates RISC-V's suitability for low-power embedded sensing applications, demonstrating that its inherent energy efficiency doesn't compromise performance in this context.
  • PM-MFM on RISC-V is Viable: The successful porting and validation of the PM-MFM method on a RISC-V platform show that complex magnetic sensing techniques can be implemented on resource-constrained, low-power architectures.
  • Analog GPIO Limitations: The study highlights that while the RISC-V system performs well, the accuracy can be slightly impacted by the use of simple analog GPIO ports compared to more sophisticated data acquisition modules found in traditional x86 setups. This points to potential areas for hardware optimization.
  • Cost-Effectiveness and Compactness: The combination of RISC-V's open-source nature and Hall sensors' inherent low cost and small size offers a pathway to highly cost-effective and compact displacement sensing solutions.

Key Results

  • The RISC-V based displacement sensor system achieved comparable linearity and accuracy to an x86 system.
  • The R² values for linear fitting were consistently high (above 0.9995) for different measurement ranges, indicating strong linear relationships.
  • The maximum relative error was less than 4.5%, aligning with findings in previous research.
  • The power consumption of the RISC-V system was measured at an average of 3.4 W, significantly lower than typical x86 systems.
  • The RISC-V platform showed only a 1% to 4% higher error compared to the x86 platform, with an average difference of 2%, largely attributed to the analog GPIO interface.

Limitations and Future Work

  • Analog GPIO Accuracy: The paper acknowledges that the accuracy of the RISC-V system is slightly reduced due to the use of a simple analog GPIO port on the Lichee Pi, compared to dedicated analog data collection modules.
  • Future Work: The authors suggest that further research could focus on optimizing the hardware interface, potentially using dedicated ADCs or more advanced signal conditioning circuits, to further improve accuracy. Exploring other RISC-V platforms and different types of Hall sensors could also expand the applicability of this approach. Additionally, investigating the integration of more complex displacement sensing algorithms or advanced power management techniques on RISC-V could be beneficial.

Summary updated

BibTeX
@misc{sun2024low,
  author = {Tao Sun and Yue Song and Huiyun Yang},
  title = {{Low-Power Magnetic Displacement Sensor Based on RISC-V Embedded System}},
  year = {2024},
  doi = {10.3390/s24134224},
}

Loading related content...

Figure