ROS robotic arm
———— ——
Map Navigation▕ voice interaction▕ Deep Vision▕ Omnidirectional motion
Product Introduction
———— ——
HunterArm is an omnidirectional mobile robot developed based on the ROS robot operating system, adopting the popular Mecanum wheel travel structure and supporting Jetson motherboard as the main control platform. The whole machine is equipped with high-performance hardware, including laser radar, depth camera, far-field microphone array, and six degree of freedom robotic arm, with excellent scalability. HunterArm can easily achieve various applications such as robot motion control, remote communication, mapping and navigation, obstacle avoidance, autonomous driving, and voice control.
In addition, HunterArm comes with over 200 pages of ROS learning courses, covering comprehensive content from beginner to advanced, and providing professional technical support to help users easily master ROS development skills and master robot applications. HunterArm is your ideal choice for learning, development, and research!
ROS operating system Lidar SLAM Deep Vision SLAM Dynamic obstacle avoidance
3D real scene mapping Machine vision Deep learning Autonomous driving
APP mapping navigation Cross platform manipulation Voice interaction Moveit simulation
Multi aircraft formation Python Handle control JetsonNano
Path planning Multi-point navigation Machine learning Identify and track
Textbook Development Manual Tensorflow Pytorch Technical Support
ROS Robot Arm Function List
———— ——
Experimental courses from introductory to advanced ROS
1 Depth Camera | |||
1) Real time video | 6) MediaPipe Development | 11) Depth following | 16) Yolo object detection |
2) QR code creation | 7) Color recognition | 12) Visual patrol line | 17) Yolo+TensorRT object recognition |
3) QR code recognition | 8) KCF target tracking | 13) Autonomous driving | 18) Lane keeping |
4) Gesture control | 9) SVM human tracking | 14) RTAB Map 3D mapping | 19) Traffic sign recognition |
5) Camera calibration | 10) TRT human tracking | 15) RTAB Map 3D Navigation | 20) Using TensorRT to accelerate flag recognition |
2 ROS controller | |||
1) VNC remote control | 6) Handle control | 11) IMU linear velocity angular velocity calibration | 16) TensorRT Acceleration |
2) NoMachine remote control | 7) Multi machine communication configuration | 12) AP hotspot mode | 17) Set AP and STA modes |
3) SSH remote control | 8) GPU acceleration | 13) Client mode | 18) Multi aircraft formation |
4) MQTT remote control | 9) Robot serial communication | 14) Real time control of rqt | 19) OLED screen display |
5) Keyboard control | 10) IMU and odometer data release | 15) GPU acceleration | 20) Ultrasonic Follow |
3 6-degree-of-freedom robotic arm | ||
1) Upper computer control | 3) Movelt simulation | 5) Movelt configuration |
2) GUI Control | 4) Kinematic control | 6) Movelt kinematic design |
4 Laser Radar | ||
1) Gmapping mapping | 6) Fixed-point navigation | 11) ROS APP mapping |
2) Hector mapping | 7) Multi point navigation | 12) ROS APP Navigation |
3) Karto mapping | 8) Path planning | 13) Multi machine navigation |
4) Cartographer mapping | 9) Automatic navigation and obstacle avoidance | 14) Multi aircraft formation |
5) Fully automated map construction | 10) Lidar tracking | / |
5 Six Channel Microphone Array | |||
1) Command wake-up | 3) Far-field pickup | 5) Voice controlled car movement | 7) Voice controlled multi-point navigation |
2) Sound source localization | 4) Noise reduction recognition | 6) Voice controlled autonomous navigation | 8) Voice controlled ultrasonic tracking |
6 ROS Robot Expansion Class | ||
1) Serial communication | 5) Low voltage alarm | 9) Control RGB light bar |
2) Control serial bus and PWM servo | 6) CAN bus communication | 10) Short circuit protection |
3) IMU attitude sensor | 7) SBUS model airplane remote control | 11) Overheating protection |
4) Battery voltage detection | 8 Control buzzer | / |
Omnidirectional movement, hand eye integration
———— ——
Mecanum wheel chassis moves in all directions
The Mecanum wheel omnidirectional chassis can help HunterArm move 360 ° in all directions, easily challenging complex routes (forward, lateral, diagonal, rotation), and move freely in all directions.
Hand eye integration, 3D space grasping
HunterArm is equipped with a six degree of freedom visual robotic arm, which can be developed to achieve functions such as autonomous grasping and sorting. Combined with LiDAR, autonomous navigation and transportation can also be achieved.
Dual brain architecture collaboration performance upgrade
———— ——
HunterArm adopts the latest dual master architecture design, with JetsonNano as the upper computer brain, leading ROS algorithm planning and visual processing. ROS Robot Extension is responsible for motion control and sensor data processing. Intelligent division of labor between the two achieves dual improvement in performance and efficiency.
① Nano main control board ② ROS Robot Expansion Board
ROS Artificial Intelligence Interaction System
———— ——
Equipped with the independently developed ROS artificial intelligence interaction system, it does not rely on command lines or complex operation methods, and can monitor the robot's status and information in real time. It also supports interactive functions such as mapping, navigation, and motion control.
Deep learning unmanned driving sandbox
———— ——
HunterArm is based on the ROS system and integrates PyTorch deep learning framework, OpenCV, YOLO object detection algorithm, and TensorRT inference engine. With the widespread application of autonomous driving in the field of new energy vehicles, it can provide convenient development support for AI autonomous driving projects.
① Road sign detection ② Traffic light recognition ③ Lane keeping ④ Patrol driving ⑤ Smart Transportation Sandbox Kit (optional)
ROS Function Introduction
———— ——
● Introduction to Lidar Functions Mapping Navigation - Exploring SLAM Algorithm |
|
01 Lidar mapping and navigation
The HunterArm robot can achieve SLAM function through LiDAR, and can develop SLAM algorithms such as Gmapping, Hector, Karto, and Cartographer for mapping. It also supports path planning, fixed-point navigation, and multi-point navigation.
02 Multi point navigation dynamic obstacle avoidance
Support fixed-point navigation and multi-point navigation, implement path planning and positioning based on ROS system, and meet the requirements of intelligent applications in multiple scenarios.
03 Dynamic obstacle avoidance in path planning
Support global path planning, multiple local path planning algorithms, real-time detection of obstacles during navigation, and re planning of paths for obstacle avoidance.
04 ROS independent exploration and mapping
Without human intervention, ROS robotic arm robots autonomously explore through algorithms to complete mapping, save maps, and return to the starting point.
05 Radar Follow
By scanning surrounding targets with LiDAR, when a target object is detected to be moving, an algorithm is used to detect the direction and distance of movement, thereby controlling the movement of the car.
● Introduction to Camera Functions 3D Vision - Visual Algorithm |
|
01 RTAB-VSLM 3D Visual Mapping and Navigation
The RTAB algorithm is used to fuse visual and radar data, achieving 3D visual mapping and navigation obstacle avoidance for robots, and supporting global repositioning.
02 Depth Image Data Point Cloud Image
Through the corresponding API, depth images, color images, and point cloud data of the camera can be obtained.
03 Opencv Machine Vision Processing
Integrate Opencv machine vision library, providing various algorithm routines, including color block detection, face detection, identity recognition, human detection, edge detection, visual inspection, etc.
① Edge detection ② Color block detection and recognition ③ Line patrol identification ④ Human body recognition
04 AR Augmented Reality By selecting the corresponding graphics, let the graphics pass through AR Enhanced technology is presented on the AprilTag tag code. | 05 KCF Target Follow The image-based KCF correlation filtering algorithm can select any object in the image and achieve target tracking. |
06 YOLO Object Recognition Based on neural networks, commonly used household items can be identified, and ROS will publish ROS topic messages based on the recognition results. | 07 Support mainstream AI deep vision Supports KNN algorithm, TensorFlow, Pytorch, YOLO, and TensorRT GPU acceleration. |
● Interconnected control/multi aircraft formation Remote Control - Cluster Control |
|
01 Multi robot Cluster Control
The keyboard can control multiple robots to complete consistent actions and achieve synchronous control
02 Remote control
MQTT enables remote control of data transmission and robot collaboration.
● Six degree of freedom robotic arm/kinematics Movelt robotic arm hand eye integration |
|
01 Movelt Robot Arm Interactive Control
Movegroup communicates with robots through ROStopics and actions to obtain their status (position, nodes, etc.). Obtain point cloud or other sensor data, and then transmit it to the robot controller (providing URDF kinematic biomimetic model).
02 ROS+Robot Arm Kinematic Visual Grasping
The kinematics plugin (The KinematicsPlugin) is used for kinematics, and one can also write their own inverse kinematics algorithm to work with the camera to achieve visual grasping of objects and other functions.
Cross platform interconnection and control
———— ——
Mobile APP control
PC control
Handle control
APP main interface
Lidar Mapping Navigation Robotic Arm Camera Overview
Map navigation interface
Product Features
———— ——
1. Mechanical arm claws | 5. 7-inch display screen | 9. Audio system | 13. Quick disassembly battery compartment |
2. Depth camera | 6. Closed shell | 10. Encoder motor | 14. Battery |
3. Bus servo motor | 7. External gain antenna | 11. Network card | 15. Auxiliary control board |
4. Main control board | 8. Laser radar | 12. Microphone array | 16. McNam Wheel |
01 High performance LiDAR
LIDARS2 LiDAR is equipped with a high-speed image processing chip, which can perform 360 degree laser ranging scanning within a radius range.
LiDAR | |||
Working Voltage | 5V | Scanning angle | 360° |
Operating current | 250mA | Output interface | UART serial port |
Laser Safety | ClassI | Operating Temperature | -10°C~+40°C |
Measure distance | 150mm-6000mm | Pitch angle | 0°~1.5° |
Angular resolution | ≤1° | Laser light source | 780nm LD |
02 Depth Vision Camera
Adopting the Obi Zhong Guangda white camera, it can obtain depth data and includes all the functions of a regular USB camera.
Deep vision camera | |||
Baseline | 40mm | Data transmission | USB2.0 Type-C |
Working distance | 0.2m-2.5m | Power consumption | Average power consumption<2.3W |
Depth accuracy | 1.0%@1m | Power supply method | Type-C |
Depth image | Supports a maximum resolution of 1024x768 | Applicable scenarios | indoor |
Depth FOV | H79°V55°D88.5°+3° | Safety | Class1 Laser |
03 High definition touch 7-inch LCD display screen
Touching the screen facilitates the display and operation of the robot, with a built-in ROS artificial intelligence interaction system that enables corresponding functions by simply touching it.
90°Foldable 7-inch HD Touchscreen LCD Screen | |||
Display Size | 7inch | Operating Temperature | Raspberry Pi/Nvidia/Windows |
Foldable maximum angle | 90° | Brightness | 500LCM |
Use | Windows/Linux | Size | 165*110*20mm |
Resolution | 1024*600PX | Weight | ≤266g |
05 6-degree-of-freedom robotic arm
Equipped with AI visual camera
30KG can only be used for bus servos
6-degree-of-freedom robotic arm | |||
Servo model | S370 High Voltage Line Bus Servo | Working Voltage | 6V-12V |
Degree of freedom | 6 degrees of freedom | Material of robotic arm | Aluminum alloy |
Mechanical arm load | 500g | Locked-rotor torque | 30kg.cm |
Protect | Locking protection, overcurrent protection, overvoltage protection, overheating protection | Mechanical arm extension | 428mm |
Product Parameters
———— ——
Product Name: | HunterArm ROS Navigation Robot Education Kit |
Product Master Control: | Jetson Nano 4GB |
Drive Expansion Board: | XR-PNL-4RL-32 |
Progressive Structure: | Mecanum Wheel Chassis |
LiDAR: | XR-LIDAR S2 LiDAR |
Camera： | DCW series depth camera |
Motor： | XR520 DC belt encoder reduction motor |
Battery： | 12.6V 2200mah |
Body material: | All aluminum alloy/anodizing treatment process |
Robotic arm： | 6-degree-of-freedom robotic arm/30kg metal axis servo |
Programming language | Python/C++ |
Voice Recognition Module: | Six microphone array module |
Product dimensions: | 346*249*340mm |
Machine vehicle self-weight: | 3460g |
Load capacity: | 3000g |
Maximum speed: | ≥ 1.2m/s (release speed limit) |
Control method: | APP、 Wireless controller, serial port, CAN, etc |
Guidelines for Small Car Experiments
———— ——
1、ROS System and Liunx Fundamentals | 7.5 Aruco Augmented Reality |
1.1 Introduction to ROS and ROS Environment Construction | 7.6 KCF Specific Target Tracking |
1.2 Introduction to ROS and ROS Environment Construction | 7.7 SVM Human Tracking |
Introduction to ROS File System | 7.8 Based on MediaPipe gesture recognition |
1.4 Introduction to Basic Commands Commonly Used in Linux | 7.9 Follow based on depth camera |
Basic use of Vim editor under Linux 1.5 | 7.10 Practice of single line visual line finding |
1.5 Experimental Summary | 7.11 ROS Robot Facial Tracking Based on gimbal |
1.6 ROS Program Creation and Compilation | 7.12 ROS Robot Color Block Tracking Based on gimbal |
Introduction to ROS Node 1.7 | 7.5 Aruco Augmented Reality |
1.8.ROS Common Basic Commands | 7.6 KCF Specific Target Tracking |
1.9. Introduction to ROS Visualization Tools | 7.7 SVM Human Tracking |
1.10. Creating ROS Messages and ROS Services | 7.8 Based on MediaPipe gesture recognition |
1.11 Writing ROS message publishing subscribers | 8、Robot deep learning |
1.12 Writing ROS Service Nodes Service and Client | 8.1 Object classification and detection based on Jetson deep learning |
1.13 URDF Model Writing and Importing | 8.2 Object detection based on YOLO v4 |
2、Introduction to ROS Robot System | 8.3 Human pose detection based on Pytorch pose |
2.1 Introduction to ROS Robot Hardware | 8.4 Gesture pose detection based on pytorch hand pose |
2.2. Instructions for unboxing and use | 8.5 Human body tracking based on TRT neural network |
2.3 Robot Source Code File Description | 9、Robot Expansion |
2.4. Instructions for Startup Service | 9.1 Radar data clipping |
2.5 XROSManager | 9.2 Radar PID Follow |
2.5.Nomachine | 9.3 Ultrasonic Follow |
3、ROS Control Fundamentals | 10、Multi robot cluster control |
3.1 Fundamentals of Mobile Control | 10.1 Naming of Multiple Robots |
3.2 Fundamentals of Servo and Handle Control | 10.2 Multi robot formation based on speed |
3.3 IM automatic calibration | 11、Remote control |
4、ROS SLAM mapping and navigation | 11.1 Remote control based on MQTT |
4.1 ROS Robot Slam Building Maps and Saving | 12、Training based on neural network dataset |
4.2 ROS Robot Explore and Build Maps | 12.1 Training of Object Detector Based on PyTorch |
4.3 ROS Robot Single point and Multi point Automatic Navigation and Obstacle Avoidance | 12.2 Classification detector based on PyTorch |
4.4 ROS robot deep binocular and LiDAR construction for 3D mapping and navigation | 12.3 Training on YOLOv4 based object detection dataset |
5、ROS Android APP | 13、Autonomous driving based on 0penCv |
5.1 Use of ROS APP | 13.1 Lane detection based on 0pencv |
5.2 ROS Robot Android/Tablet APP Mapping | 13.2 Lane Keeping |
5.3 ROS Robot Android/Tablet APP Navigation | 13.3 Traffic Sign Detection Based on Neural Networks |
6、Speech recognition | 13.4 Integration of Lane Keeping and Object Detection |
6.1 Introduction and Parameter Configuration of Six Array Microphone | 14、Neural network-based autonomous driving |
6.2 Robot Speech Recognition Solution | 14.1 Driving data collection |
6.2 Voice wake-up and sound source localization | 14.2 Driving data training |
6.3 Voice Control and Navigation | 14.3 Use trained models for inference |
7、Robot Vision | 14.4 Traffic Sign Hybrid Neural Network for Autonomous Driving |
7.1 Using HSV tool to search for HSV color values | 15、Based on the use of Moveit robotic arm |
7.2 Color Block Recognition Practice | 15.1 Use of Moveit robotic arm on virtual machine |
7.3 Facial Recognition and Identity Recognition Practice | 15.2 Create your own Moveit workspace |
7.4 Edge detection | / |
Shipping List
———— ——
Advanced version car Professional version car Charger Handle Screwdriver
| Optional robot specific aviation aluminum box Optional robot aviation aluminum box for shipment, the box body is not easily deformed, sturdy and durable, ensuring safety and protection during long-distance transportation. If needed, please contact customer service for price difference and optional selection |
Our services
———— ——
About express delivery | Gift materials | Technical Support |
Our goal is to continuously improve our own technical capabilities and place greater emphasis on the customer's product experience.
