PHCG: Optimizing Simulink Code Generation for Embedded System With SIMD Instructions

Su, Zhuo; Wang, Dongyan; Yu, Zehong; Yang, Yixiao; Jiang, Yu; Wang, Rui; Chang, Wanli; Li, Wen; Cui, Aiguo; Sun, Jiaguang

doi:10.1109/tcad.2022.3197504

Cited by 5 publications

(6 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This software facilitates the modeling, simulation, and code generation for embedded control software. Consequently, its capability for automatic code generation significantly reduces the manual coding workload for developers, as emphasized in a previous study [3].…”

Section: Introductionmentioning

confidence: 87%

“…Moreover, it provides the advantage of incrementally testing each sub-process throughout the design phase [2]. SIMULINK, acknowledged as a prominent model-driven design tool, has gained significant popularity in embedded applications, particularly in sectors like smart transportation, avionics, and vehicles, as indicated by citations [3][4][5]. This software facilitates the modeling, simulation, and code generation for embedded control software.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Rapid Experimental Protocol for PMSM via MBD: Modeling, Simulation, and Experiment

Hu,

Ahn,

Kang

et al. 2024

Computers

View full text Add to dashboard Cite

As control algorithms evolve, their enhanced performance is often accompanied by increased complexity, reaching a point where practical experimentation becomes unfeasible. This situation has led to many theoretical studies relying solely on simulations without experimental verification. To address this gap, this study introduces a rapid experimentation protocol (REP) for applying field-oriented control (FOC) strategies to permanent magnet synchronous motors (PMSMs) based on model-based design (MBD) and automated code generation. REP is designed to be user-friendly and straightforward, offering a less complex and more accessible alternative to DSP toolboxes. Its excellent hardware compatibility is conducive to code porting and development. With this protocol, users can quickly conduct FOC strategy experiments with reduced dependency on the complex automated code generation tools often associated with toolboxes. Centered around the PMSM model, this method utilizes only the fundamental modules of MATLAB2023b/Simulink, greatly simplifying the user experience. To demonstrate the feasibility and efficiency of the protocol, models for both sensor-based and sensorless control are developed. The practicality of REP, including sensor-based and sensorless experiments, is successfully validated on an arm-cortex-M4-based GD32 microcontroller.

show abstract

Section: Introductionmentioning

confidence: 87%

Section: Introductionmentioning

confidence: 99%

Rapid Experimental Protocol for PMSM via MBD: Modeling, Simulation, and Experiment

Hu,

Ahn,

Kang

et al. 2024

Computers

View full text Add to dashboard Cite

show abstract

“…For two tasks i and k without dependency (with parallel relation), task i begins either earlier or later than k, as shown in (17). If they are mapped onto the same SPE.…”

Section: A Formulation Of One-step Ilpmentioning

confidence: 99%

Task Mapping and Scheduling on RISC-V MIMD Processor With Vector Accelerator Using Model-Based Parallelization

Wu,

Kumano,

Marume

et al. 2024

IEEE Access

View full text Add to dashboard Cite

In this paper, we propose a model-based workflow to generate parallel code on a multiple instruction stream, multiple data stream (MIMD) processor with vector accelerator (MIMDV) from a Simulink model. Solving data-and task-parallelism is crucial during this process. For data parallelism, a RISC-V Simulink library written in vector codes was prepared for blocks with sufficient vector or matrix calculations. Moreover, large inputs can be divided, which means that tasks can be executed simultaneously using multiple cores. For task parallelism, integer linear programming (ILP) was designed to deploy tasks on scalar processing elements (SPEs) and a vector processing element (VPE) of MIMDV. The use of a vector library for a task and the number of SPEs a task uses were determined. To reduce the overhead, synchronization is realized by barrier wait, and execution is divided into multiple time intervals called layer. We propose a novel one-step ILP that accurately minimizes the parallel time of such a situation. Furthermore, we propose a two-step ILP to achieve reasonable performance in practical time. One step is SPE mapping, and the other is layer scheduling. We tested our methods using random task graphs and real-world applications on DR1000C, a type of RISC-V MIMD processor with a vector accelerator.

show abstract

“…Furthermore, it is advantageous in that it facilitates the incremental testing of each subprocess throughout the design stage [8]. MATLAB/Simulink, a prominent model-driven design tool, has gained substantial popularity in embedded applications, particularly in sectors such as smart transportation, avionics, and vehicles, as evidenced by references [9][10][11]. This tool facilitates modeling, simulation, and code generation for embedded control software.…”

Section: Introductionmentioning

confidence: 99%

“…This tool facilitates modeling, simulation, and code generation for embedded control software. Consequently, its capacity for automatic code generation significantly alleviates the burden of manual coding for developers [9].…”

Section: Introductionmentioning

confidence: 99%

Novel rapid control prototyping for permanent-magnet synchronous motor via model-based design and STM32 chip

Hu,

Ahn,

Kang

et al. 2024

Preprint

View full text Add to dashboard Cite

As control algorithms become increasingly sophisticated, delivering improved performance at the expense of greater complexity, practical experiments are often unfeasible. To address this issue, this study devised a novel rapid control prototyping (NRCP) based on model-based design (MBD) by MAT LAB/Simulink, STM32CubeMX software, and field-oriented control (FOC) strategies for permanent-magnet synchronous motors (PMSMs). The NRCP avoids complex toolboxes and utilizes only MATLAB/Simulink’s model-to-C language tool, facilitating code portability and development. By using the NRCP, users can quickly experiment with FOC strategies, reducing their reliance on complex automatic code generation tools typically associated with toolboxes. This study focused on the PMSM model, employing only the basic modules of MATLAB/Simulink to generate the application-layer code and the STM32CubeMX software to do low-level code generations, significantly simplifying user experience. To demonstrate the feasibility and efficiency of the protocol, sensor-based and sensorless control models were developed using the MBD method. Finally, the practicality of the NRCP was successfully validated via sensor-based and sensorless experiments using an ARM Cortex-M4-based STM32 microcontroller.

show abstract

PHCG: Optimizing Simulink Code Generation for Embedded System With SIMD Instructions

Cited by 5 publications

References 28 publications

Rapid Experimental Protocol for PMSM via MBD: Modeling, Simulation, and Experiment

Rapid Experimental Protocol for PMSM via MBD: Modeling, Simulation, and Experiment

Task Mapping and Scheduling on RISC-V MIMD Processor With Vector Accelerator Using Model-Based Parallelization

Novel rapid control prototyping for permanent-magnet synchronous motor via model-based design and STM32 chip

Contact Info

Product

Resources

About