Recently, DFRobot has developed an exhibit of a line-following robot, which uses a RISC-V control board equipped with the JH7100 chip from Seaflow Technology and runs the Fedora Linux operating system. With the support of the PinPong Python library, GPIO control is achieved using the Mind+ programming software. This robot utilizes the HUSKYLENS (AI visual recognition) sensor for intelligent route identification, enabling automatic line-following functionality.
1
Exhibit Description
1. Seaflow Technology JH7100
The JH7100 is equipped with a dual-core U74, sharing a 2MB L2 cache, with a working frequency of up to 1.2GHz, supporting Linux operating systems. The ISP developed by Seaflow Technology adapts to mainstream sensors, and the built-in image and video processing subsystem supports H265/H264/JPEG encoding and decoding. The integration of high-performance, low-power Vision DSP and NNE makes application development smarter and more efficient. The JH7100 can handle various complex video image processing and intelligent visual computing tasks, meeting the real-time processing needs of edge devices.
2. PinPong Library
The PinPong library is a Python library for controlling open-source hardware control boards, based on the Firmata protocol and compatible with MicroPython syntax, allowing users to quickly start controlling open-source hardware with Python in just 5 minutes. With the PinPong library, you can program various common open-source hardware directly using Python code. The principle is to burn a specific firmware into the open-source hardware, enabling it to communicate with a computer via serial port and execute various commands. The design of the PinPong library aims to free developers from the constraints of complex hardware models during development, allowing them to focus on software implementation. Even if the program is initially developed using Arduino, it can be seamlessly switched to a control board during deployment by simply modifying the hardware parameters to ensure normal operation, achieving the goal of “write once, run anywhere”. With the joint efforts of engineers from DFRobot and Seaflow Technology, the PinPong library has been fully adapted to the Seaflow RISC-V control board, allowing smooth operation to read various sensors and control actuators directly.
3. Mind+
Mind+ is a domestically developed programming software for teenagers with independent intellectual property rights, integrating various mainstream control boards and hundreds of types of open-source hardware, supporting artificial intelligence (AI) and Internet of Things (IoT) functionalities. It allows both graphical block programming and the use of advanced programming languages such as Python/C/C++, enabling everyone to easily experience the joy of creation. Mind+ supports both graphical programming and Python programming, simplifying environment configuration and other tasks to achieve a ready-to-work state immediately. Currently, DFRobot has completed the adaptation of the Seaflow Technology JH7100 SoC, allowing low-code development for the Seaflow RISC-V control board, further reducing the development difficulty of the Seaflow VisionFive single-board computer.
4. HUSKYLENS (AI Visual Recognition) Sensor
This line-following robot uses the HUSKYLENS (AI visual recognition) as the vision sensor. HUSKYLENS is a user-friendly AI camera (vision sensor) with six built-in functions: face recognition, object tracking, object recognition, line tracking, color recognition, label recognition, object classification, QR code recognition, and barcode recognition. AI training can be completed with just one button, eliminating tedious training and complex visual algorithms, allowing developers to focus more on project conception and implementation. HUSKYLENS features onboard UART/I2C interfaces, allowing seamless connection to mainstream development boards such as Arduino, Raspberry Pi, LattePanda, and Microbit, directly outputting recognition results to the controller without the need for complicated algorithms, enabling the creation of very innovative projects. This robot exhibit achieved compatibility between HUSKYLENS and the Seaflow Technology RISC-V control board.
Using HUSKYLENS is very simple. You can switch algorithms by toggling the function button; pressing the learning button allows the AI visual recognition to learn new things. It has a built-in 2.0-inch IPS display, which allows parameter adjustments without a computer. The debugging process and recognition results are displayed directly on the screen, providing a WYSIWYG experience. The built-in machine learning technology enables face recognition and object recognition capabilities. By long-pressing the learning button, it can learn from different angles and ranges, becoming more accurate with each learning session.
2
Implementation Steps
1. Functionality
a) Use PinPong to drive the robot and use AI visual recognition for line following;
b) The AI visual recognition communicates with the Seaflow JH7100 via I2C bus;
c) Use 2 IO ports to control motor direction;
d) Use 2 PWM ports to control motor speed;
e) Automatically run on startup.
2. Features
a) Automatic operation;
b) Expandable development;
3. Usage Steps
a) After powering on the mainboard, it will immediately execute /root/linefollower.py;
b) The line-following recognition learning needs to be controlled manually. Click the button in the upper right corner of HUSKYLENS to learn the black line that needs recognition, and when the white arrowed line turns blue, it indicates that learning is complete;
c) After learning is complete, it can be placed on the line-following track to start following the line.
This case demonstrates the enormous potential of Seaflow Technology’s RISC-V development board in the field of robotics and its ability to adapt well to various sensors for IoT control. In the next phase, Seaflow Technology will collaborate with DFRobot to develop more IoT industrial control applications based on the Seaflow VisionFive single-board computer and the new generation JH7110 chip, promoting the application of IoT in practice.