AI Discipline
Lower Deployment Threshold for Multimodal AI Large Models, Enhance On-Device Training Efficiency, and Provide Comprehensive Support from Model Implementation to Application Validation
OS-nano
OS-nano
Powered by NVIDIA AI, OS-nano fuses visual and voice perception for environment understanding and enables autonomous task execution for a wide range of embodied robot applications.
Multi-scenario Fusion Navigation Algorithm
Fusion of multi-source sensing data and diverse algorithms supports navigation development and application in complex environments, laying a solid technical foundation for autonomous driving, robotics and enhanced navigation applications.
Multi-stage Practice and Extended Functions
A full-process workflow covering hardware, algorithms, system integration, installation, and innovative research — designed to drive practical innovation and solve complex technical problems.
- SLAM & AI Practice ：Localization, Planning, Navigation & AI Control Synergy
- Deep Model Training ：Visual Annotation, Simulation & Pre-trained Model Training
- AI Large Model Dev ： AI Algorithm Expansion for Diverse Scenarios
- Hardware Assembly & Debugging ：Chassis Hardware & Electrical Commissioning
- Motion Control Calibration ：Chassis CAN Bus & Control Calibration
- Chassis-Nav Joint Debugging ：Multi-sensor Calibration & Navigation Tuning
Multi-domain Application & Development
Powered by modular hardware, multimodal AI large-model deployment, and ROS ecosystem integration, OS-nano delivers an all-in-one robotics platform featuring AI algorithms, embodied interaction, autonomous navigation, and vehicle–road collaboration.
-
-
Embodied AI Discipline
Hardware Entity + Multimodal Perception + Real-Time Action Control; Key support for agent-environment interaction intelligence development
-
Mobile Robotics Discipline
Navigation Module & Modular Chassis Building. Full-cycle support from basic learning to in-depth innovation for autonomous navigation, scenario adaptation and function expansion.
-
Intelligent Connected Discipline
Enable the implementation of vision-based autonomous driving, advance the vehicle-environment interaction capability, and provide end-to-end full technical development solutions.
Modular UGVs Expansion
Configuration & Functions
Comprehensive Application Development
-
Voice Control
-
Lane Recognition
-
Swarm Control
-
Visual Navigation
Multi-industry Applications
-
Development
-
Training
-
Education
-
Research
Support diverse modular applications and work with you to unlock new possibilities across industries.
Parameter
- Available on：OMNI UGV, Ackermann steering UGV, Differential Steering UGV
- Operating Environment：Indoor / outdoor（non-rainy and non-snowy）
- Operation System：Linux Ubuntu 22.04 ROS 2
- Embedded Computer：Jetson Orin nano （8G）
- Storage：128G
- Computing Power：40 TOPS
- Power：10 W
- LiDAR：2D LiDAR(16m Measuring Radius)
- Depth Camera：Astra Pro Plus
- Sound Pick-up Unit：six-channels microphone, 10m detecting range, built-in neural network processor, custom voice lexicon
- Hardware Functions：Hardware Installation & Commissioning, CAN Communication Analysis & Application, Chassis Kinematics Analysis & Application, Battery BMS Analysis & Application
- Auto-Nav Install, Debug & Test Function：Auto-Nav Hardware Install, Debug & Test, Auto-Nav Software Architecture Analysis & Application, Sensor Fusion Data Analysis and Application, Chassis Control Fusion Analysis and Application
- Nav Functions：Dynamic Obstacle Avoidance, Point-to-Point Navigation, Multi-Point Navigation, TEB and DWA Path Planning, LiDAR Angle Masking, LiDAR Mapping Navigation
- Mapping Functions：Rtab Visual-Only/Rtab Visual-LiDAR /Gmapping/Hector/Karto/Cartographer Mapping, RRT/Frontier/Explore_Lite Autonomous Mapping
- Robot Formation Function： Leader-Follower Algorithm
- HMI Function ：Sound Source Localization, Voice Summon/Control/Navigation/Broadcast/Interaction, LiDAR Following
- Vision Function ：OpenCV Application and Tutorial, Webcam Monitor, Depth Vision Follow, KCF Follow, AR Tag Recognition, RGB Line Patrolling, Human Recognition & Following, 3D Vision Mapping, 3D Vision Nav
- Deep Learning Function：YOLO Object/Traffic Sign Recognition, Deep Learning Model Training, Gesture Contro, Sandbox Map Auto-Driving
- AI Large Model Deployment：Large Model Application/Semantic Understanding/Emotion Perception/Voice Interaction
- AI Vision Large Model：Vision Large Model Target Recognition/Scene Understanding/Text Recognition
- AI Large Model Robot：Multimodal Large Model Voice Control/Auto Line Patrolling/Color Tracking, Embodied Intelligence Real-time Monitor/Vision Tracking/Smart Butler, Multimodal Information Fusion Overview
- Configuration：Nav Module + Chassis
- Configuration Description：Components and functions support modular upgrades; specifications are subject to the actual product configuration.
- Technical Documents：API Interface, Open-Source Code, Chassis CAN Protocol, Chassis STP Drawing, ROS Driver Package
- Services：Online Technical Support
Technical Support
