Abstract: The Cortex-M architecture is designed around the concept of low-power, real-time control, with the advantage of being able to tightly couple I/O and peripherals through a very lightweight software layer. However, this architecture is not ideal for tasks that process large amounts of data, and due to the increasing complexity of motor control algorithms, the computational performance requirements for local data analysis are also growing, which poses some challenges to microcontroller unit (MCU) with Cortex-M architecture. In order to solve this problem, this paper designs a new hardware accelerator, which uses the simplicity and efficiency of the CORDIC algorithm to achieve high-precision trigonometric function operation, and uses the hardware module to realize the entire control process of field oriented control (FOC), and the maximum delay for completing a FOC operation is 264.5 ns. This drastically reduces the runtime and energy consumption of signal processing algorithms, and the new accelerator module will provide users with more precise and complex running algorithms while maintaining energy efficiency.

Key words: FOC control; hardware accelerators; CORDIC algorithm; motor control

摘要： Cortex-M 架构围绕低功耗、实时控制的概念而设计，其优势在于能够通过非常轻量级 的软件层来紧密耦合地控制 I/O 和外围设备。然而这种架构对于处理大量数据的任务并不理 想，由于电机控制算法的复杂性增加，对本地数据分析的计算性能要求也在增长，这对 Cortex-M 架构的微控制单元（MCU）带来了一些挑战。为解决这一问题，文章设计出一款新 型硬件加速器，利用 CORDIC 算法的简单高效性实现高精度的三角函数运算，并用硬件化模 块实现磁场定向控制（FOC）的整个控制流程，完成一次 FOC 运算过程的最大延时为 264.5 ns。 这大幅减少了信号处理算法的运行时间和能量消耗，新加速器模块将为用户提供更精确、更 复杂的运行算法，同时保持能源效率。

关键词: FOC 控制；硬件加速器；CORDIC 算法；电机控制