1 5

Yobot

Machine vision kit

Machine vision kit

$2,900.00 USD
$2,900.00 USD
Giảm giá Đã bán hết

Product Name: Machine Vision & Deep Learning Experimental System

Positioning: An open-source, full-stack educational platform combining "Vision + AI + Control," tailored for universities and vocational institutions.

▍ Product Overview: This system is a comprehensive machine vision training platform designed to bridge the gap between software algorithms and physical execution. Moving beyond traditional data-only image processing, it utilizes standard vision hardware like area scan cameras and features 18 built-in classical vision experiments. By integrating a robotic arm and the TensorFlow deep learning framework, learners can use AI algorithms to drive the robotic arm for real-time object grasping and sorting—transforming visual data into physical action.

▍ Core Functions

  • Comprehensive Vision Training: Includes 18 practical experiments covering general image processing, giving users a complete understanding of visual inspection systems.

  • Industrial Hardware Selection: Provides detailed guides on selecting cameras, lenses, and light sources, equipping students with practical industrial engineering skills.

  • Hardware-Software Synergy: Integrates a robotic arm to expand visual application scenarios, moving from simple image processing to physical interaction and sorting.

▍ Key Advantages & Educational Value

  1. Fully Open-Source for Secondary Development: Built on the Open-CV platform, the software and application codes are completely open-source, offering maximum flexibility for academic research and custom development.

  2. Lowering the AI Threshold: Seamlessly introduces the TensorFlow deep learning framework, making complex AI algorithms accessible to learners.

  3. Rapid Learning Curve: Comes with fully integrated experimental textbooks and instructional videos, ensuring users can quickly master the system and expand their engineering mindset.

Xem toàn bộ chi tiết

Video

Specification parameters

1) Base/Chassis Parameters:
Motor reduction ratio: ≥1:2； Load capacity: ≥12kg； Maximum speed: ≥2.7 m/s； Body weight: ≥6 kg； Dimensions: ≥400mm*500mm； Suspension system: ≥coaxial swing suspension； Off-road wheels: ≥8 inches； Battery capacity: ≥5000mAh； Battery life: ≥6.5 hours (idle) and ≥5.5 hours (3kg load)； Motor: DC Brushed Motor； Encoder: AB-phase high-precision encoder； Features: App control supports navigation, map creation, image transmission, obstacle avoidance, and other functions； Support: Supports controllers, with reserved Usart/CAN communication control interfaces and multiple IO interfaces； Operating System: The STM32 board runs on freeRTOS ROS with Ubuntu 18.04 and Melodic

2) Host computer processing unit:
GPU: 384-core NVIDIA Volta ™ GPU with 48 Tensor Cores； CPU: A 6-core NVIDIA Carmel ARM® v8.2 64-bit processor with 6MB L2 and 4MB L3 cache； Visual Accelerator: 7-Channel VLIW Visual Processor； Deep Learning Accelerator: 2 NVDLA Engines； Video memory: 8GB； Video encoding: 2x 4K60|4x 4K30|10x 1080p60|22x 1080p30 (H.265)； Video decoding: 2x 8K30|6x 4K60|12x 4K30|22x 1080p60|44x 1080p30 (H.265)； Cameras: 2 MIPI CSI-2 D-PHY channels； Connectivity: Gigabit Ethernet, WiFi； Display: HDMI and DP； USB: 4xUSB3.1, USB2.0 Micro-B； I/O: GPIO, I2C, I2S, SPI, UART

3) Lower-level machine processing unit:
CPU: Based on STM32； Core width: ≥32 bits； Maximum clock frequency: ≥72MHz； Communication modes: Supports CAN, I2C, IrDA, LIN, SPI, UART/USART, and USB； Peripheral devices: DMA, motor control PWM, PDR, POR, PVD, PWM, temperature sensor, WDT, crystal oscillator； Input/output count: ≥51； Program memory capacity: ≥256KB； Program memory type: FLASH； RAM capacity: ≥48K； Voltage-Source (Vcc/Vdd): 2 V to 3.6 V； Oscillator type: internal

4) Laser navigation scanning unit:
Range: 0.15-12m； Scan angle: 0-360°； Range resolution: <0.5mm； Angle resolution: ≤1°； Single measurement time: ≤0.5ms； Measurement frequency: ≥4000 Hz； Scan frequency: 1-10 Hz