RISC-V 機械手臂控制系統晶片軟硬體共同設計;Co-Design of Software and Hardware for RISC-V Robotic Arm Control System On a Chip

NCUIR > College of Electrical Engineering & Computer Science > Executive Master of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/95413

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/95413

Title:	RISC-V 機械手臂控制系統晶片軟硬體共同設計;Co-Design of Software and Hardware for RISC-V Robotic Arm Control System On a Chip
Authors:	周宏穎;CHOU, HUNG-YING
Contributors:	資訊工程學系在職專班
Keywords:	機械手臂;擴充指令;RISC-V
Date:	2024-07-18
Issue Date:	2024-10-09 16:47:08 (UTC+8)
Publisher:	國立中央大學
Abstract:	隨著工業自動化和智能製造的快速發展，高效、靈活且成本效益高的機械手臂控制系統需求不斷增長。然而，現有商業控制系統價格高昂且靈活性不足，難以滿足多樣化和快速變化的市場需求。因此本研究旨在開發一個基於 RISC-V 處理器的多軸機械手臂控制系統晶片開發平台，以實現智慧型機械手臂系統的快速開發和佈署，並提出一種基於 RISC-V 開放指令集架構的解決方案，設計並實現多軸運動控制擴充指令，這些指令能夠高效地驅動多軸運動控制硬體加速器，顯著提升系統效能。為提高開發效率，本研究使用 PLCopen 的 Beremiz 編輯器為中介軟體，將運動控制邏輯轉換為 XML 格式的參數表，並由 GVM 虛擬機進行解析和執行。在整體系統架構設計上，本研究採用 MIAT 系統設計方法論，通過模組化設計提高系統的靈活性和可擴展性。實驗結果表明，所開發的系統在使用 RISC-V 處理器的 FPGA 平台上執行多軸同步控制時，相較於傳統的 ESP32 平台，效能提高了 1.75 倍。這證實了系統在運動控制的精確度、響應時間和整體效能方面具有顯著優勢。本研究不僅提供了一個高效、靈活且低成本的解決方案，還展示了 RISC-V 架構在機械手臂控制領域的巨大潛力和應用前景。;As industrial automation and intelligent manufacturing rapidly advance, the demand for efficient, flexible, and cost-effective robotic arm control systems continues to grow. However, existing commercial control systems are expensive and lack flexibility, making it difficult to meet the diverse and rapidly changing market needs. Therefore, this study aims to develop a multi-axis robotic arm control system chip development platform based on the RISC-V processor, enabling rapid development and deployment of intelligent robotic arm systems. The study proposes a solution based on the RISC-V open instruction set architecture, designing and implementing multi-axis motion control extension instructions. These instructions can efficiently drive multi-axis motion control hardware accelerators, significantly enhancing system performance. To improve development efficiency, this study uses the PLCopen Beremiz editor as middleware to convert motion control logic into XML format parameter tables, which are parsed and executed by the GVM virtual machine. In the overall system architecture design, this study adopts the MIAT system design methodology, improving system flexibility and scalability through modular design. Experimental results show that the developed system, when executing multi-axis synchronous control on an FPGA platform using a RISC-V processor, achieved a 1.75 times performance improvement compared to traditional ESP32 platforms. This demonstrates significant advantages in motion control accuracy, response time, and overall performance. This study not only provides an efficient, flexible, and low-cost solution but also showcases the tremendous potential and application prospects of the RISC-V architecture in the field of robotic arm control.
Appears in Collections:	[Executive Master of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	83	View/Open

社群 sharing

Loading...