Sundance at the 49th Intelligent Sensing Program
Download as PPTX, PDF1 like595 views
Sundance Multiprocessor Technology Ltd. has developed the VCS-1 system, which uses a Xilinx Zynq MPSoC for high-performance, low-power applications compatible with a wide range of sensors. It supports multiple programming environments and frameworks, including Linux, ROS, and MQTT, and is tailored for computer vision and embedded AI tasks. Future research opportunities focus on advanced topics like object detection, autonomous navigation, and robotics integration.
![HARDWARE FEATURES
Connectivity:
FM191-R; FMC-LPC to:
15x Digital I/Os [DB9]
12x Analogue Inputs [DB9]
8x Analogue Outputs [DB9]
1x Expansion [SEIC]
FM191-U; SEIC to:
4x USB3.0 [USB-c]
28x GPIO [40-pin GPIO]
FM191-A1; 40-pin GPIO
28x GPIO [DB9]
5FM191
I/O interfaces](https://image.slidesharecdn.com/sundancevcs-49thintelligentsensingprogram-sensinginprecisionagriculture-19-09-2019-190920160249/85/Sundance-at-the-49th-Intelligent-Sensing-Program-5-320.jpg)
Sundance at the 49th Intelligent Sensing Program
- 1. 49th Intelligent Sensing Program Sensing in Precision (Digital) Agriculture Sundance Multiprocessor Technology Ltd.
- 2. THE COMPANY Established in 1989 by Flemming CHRISTENSEN • Employee Owned and a ’Life-Style’ company • 12x people with 300+ years experince • 4x with accredited Xilinx FPGA training • Always designed and built our own products • BSI - ISO9001-2015 certified, since 2003 Techology Focus • Acceleration & Vision, • Sensor & Robotics 2
- 3. • Overview • Hardware features • Sensors compatibility • Industry compatibility • Open software and firmware • PYNQ (Python) support • EDGE AI support • Applications • Endorsement • Summary • What Next? 3 OVERVIEW OF THE VCS-1 SYSTEM
- 4. A Xilinx Zynq MPSoC is the ‘heart’ of VCS-1 and provide 64-bit processor scalability while combining real-time control with soft and hard engines for graphics, video, waveform, and FPGA acceleration, using a Trenz TE0820 “SoM”. HARDWARE FEATURES 4 Multiprocessor system ZU4EV TEO820-ZU4EV
- 5. HARDWARE FEATURES Connectivity: FM191-R; FMC-LPC to: 15x Digital I/Os [DB9] 12x Analogue Inputs [DB9] 8x Analogue Outputs [DB9] 1x Expansion [SEIC] FM191-U; SEIC to: 4x USB3.0 [USB-c] 28x GPIO [40-pin GPIO] FM191-A1; 40-pin GPIO 28x GPIO [DB9] 5FM191 I/O interfaces
- 6. A custom enclosure was specially designed for accommodating the VCS-1 system, called PC104-Blade 6 HARDWARE FEATURES Rugged Enclosure
- 7. The VCS-1 is compatible with a wide range of sensors, like: Depth sensor. Interface USB3.0 Intel RealSense - D435 SENSORS COMPATIBILITY 7 Tracking sensor. Interface USB3.0 Intel RealSense – T265
- 8. INDUSTRY COMPATIBILITY VCS-1 features: Raspberry Pi version 3 model B and Arduino compatible Compatible with most of the Arduino/Raspberry Pi sensors and actuators USB3.0 & Ethernet ports for interfacing with a wide range of sensors 8
- 9. HARDWARE DEBUG TOOLS 9 VCS hardware debugging is easy with industry standard JTAG directedly into the Zynq MPSoC TULIPP’s LynSyn Power Monitor & Toolchain http://bit.ly/VCS_Lynsyn http://bit.ly/VCS_Tulipp_Toolchain Xilinx JTAG Platform Hardware Cable http://bit.ly/VCS_JTAG_Cable Xilinx JTAG SmartLynq Cable http://bit.ly/VCS_SmartLynq_Cable
- 10. OPEN SOURCE Open Source and online documentation: CERN’s Open Hardware Repository http://bit.ly/VCS_OHR VCS Hardware http://bit.ly/VCS_Hardware VCS Software http://bit.ly/VCS_Software VCS Firmware http://bit.ly/VCS_FM191_firmware 10
- 11. VCS-1 DEVELOPMENT TOOLS VCS’s “Processing System” (PS) running within the ARM A53 CPUs, is supported by Linux: ► http://bit.ly/VCS_Xilinx_Linux ► https://ubuntu.com/ VCS has a version of Ubuntu 16.04 LTS and 18.04 LTS ported and for free download. Xilinx also offers free option of an embedded Linux for Zynq, called PetaLinux: ► http://bit.ly/VCS_Peta_Linux The VCS’s “Programable Logic” (PL) – that is the ‘classic’ FPGA fabric - is supported by the free WebPACK & SD-SoC environment from Xilinx: ► http://bit.ly/VCS_Vivado_Zynq ► http://bit.ly/VCS_C_to_VHDL_SDSoC 11
- 12. PROGRAMMING TOOLS MQTT compatible OASIS standard Machine-to-machine (M2M)/"Internet of Things" connectivity protocol. OpenCV compatible Open source computer vision and machine learning software library. ROS compatible The Robot Operating System (ROS) is a set of software libraries and tools that helps building robot applications. ROS2 ready PYNQ compatible PYNQ is an open-source project from Xilinx that is utilizing Python & Jupyter to make it easy to design embedded systems with Zynq 12
- 13. The Xilinx Edge-AI Platform provides comprehensive tools and models which utilize unique deep compression and hardware-accelerated Deep Learning technology. The platform provides efficient, convenient and economical inference deployments for embedded-CPU-based FPGAs. The Xilinx AI team consists of renowned researchers and experienced professionals known for their pioneering work in the field of deep learning. 13 EDGE-AI Xilinx Edge - AI The VCS-1 is fully compatible with the Xilinx Edge-AI.
- 14. Key Features Provides a complete set of toolchains with compression, compilation, deployment and profiling. Supports mainstream frameworks and the latest models capable of diverse deep learning tasks Provides a lightweight standard C/C++ programming API (no RTL programming knowledge required) Scalable board support from cost-optimized to performance-driven platforms Supports system integration with both SDSoC and Vivado DNNDK consists of: DEep ComprEssioN Tool (DECENT) Deep Neural Network Compiler (DNNC) Neural Network Runtime (N2Cube) Profiler 14 EDGE-AI Xilinx-DPU DNNDK™ (Deep Neural Network Development Kit)
- 15. 15 EDGE-AI Xilinx Edge-AI Frameworks Xilinx Edge-AI Platform supports a number of industry-standard frameworks • TensorFlow is an open-source framework developed by Google. • CAFFE is an open-source framework developed at UC Berkley. • Darknet is an open-source framework developed by Joseph Redmon. • Keras is a high-level neural networks API, written in Python and capable of running on top of TensorFlow.
- 23. SUMMARY The VCS-1 has the following characteristics: 1. High performance, using ‘state-of-the-art’ Xilinx Zynq MPSoC 2. Low power consumption (typical 10W) 3. Highly compatible with a wide range of commercially available sensors and actuators 4. Highly optimised for computer vision applications, embedded AI and Deep Learning 5. Fully reconfigurable and expandable using PC/104 form-factor (90mm x 96mm) 6. Compatible with the most common Linux distros (I.e. Ubuntu, Debian, RedHat, CentOS, etc.) 7. Software support for ROS and MQTT 8. Python and C/C++ support 23
- 24. WHAT’S NEXT ? Research topics: Object detection, changes detection, visual-servoing Autonomous navigation, object grasping 2D>3D Pose estimation, perception, AI in robotics Sundance University Program (SUP) Access to hardware prototypes Advisory Board Members BSC/MSc/PhD Internships R&D Project proposals H2020, Innovate-UK, EuroSTARS EPSRC/BBSRC/UKRI/SPRINT https://www.smartagrihubs.eu/portal/network Want to know more about Sundance in R&D Projects ? https://www.sundance.com/sundance-in-eu-projects-programs/ 24
- 25. WHAT’S NEXT ? 25
- 26. WHAT’S NEXT ? 26
- 27. 27 Sundance Multiprocessor Technology, Ltd. Pedro Machado Pedro.M@sundance.com Flemming Christensen Flemming.C@sundance.com www.SUNDANCE.com