Resources:
Advanced On Device AI Vision Sensor With 6 TOPS Acceleration For Robotics And Intelligent Systems
Gravity HUSKYLENS 2 is a next generation AI vision sensor designed to make powerful machine vision accessible for education prototyping and real world intelligent devices. Powered by a 1.6GHz dual core processor with an integrated 6 TOPS AI accelerator it runs all vision algorithms fully on device delivering high speed low latency results without relying on cloud computing.
HUSKYLENS 2 stands out with rich AI model support flexible LLM interaction and broad controller compatibility. It includes more than twenty built in ready to use vision algorithms and also supports deployment of custom trained models. Through its built in MCP service it can both call large language models and provide real world visual perception data to them. With a replaceable 2MP camera a 2.4 inch IPS touchscreen microphone speaker and status indicators it is ideal for educational robotics interactive installations and intelligent device development.
Key Features
- High Performance On Device AI Processing
Dual core 1.6GHz processor with integrated 6 TOPS AI accelerator for real time vision tasks
- Rich Built In AI Model Library
Includes twenty plus ready to use vision algorithms with support for custom model deployment
- Flexible LLM Interaction Via MCP
Can interface with large language models and provide real world perception data
- All In One Vision Sensor Hardware
Integrated camera display microphone speaker and indicator lights
- Broad Controller Compatibility
Works seamlessly with popular microcontrollers and development platforms
Blazing-Fast On-Device AI with 6 TOPS of Power
At the core of HUSKYLENS 2 is a dual-core 1.6GHz processor with an embedded 6 TOPS AI acceleration unit. This significant computing capacity, along with 1GB of RAM, allows the HUSKYLENS 2 to run complicated algorithms and neural networks directly on-board. This enables real-time processing for challenging applications such as multi-object identification and complex pose estimation, enabling high-speed, low-latency performance without the need for cloud computing.
Start Creating Instantly with Over 20 Built-in AI Models
HUSKYLENS 2 is intended for quick productivity. It comes pre-loaded with a comprehensive library of over 20 AI models that address a wide range of popular computer vision applications, including face recognition and object tracking. This large collection enables users to start constructing projects right away with minimum preparation, drastically lowering the entrance barrier for AI application development.
Empowering LLMs with Contextual Awareness via MCP Service
HUSKYLENS 2 is the first AI vision sensor with a built-in MCP service, which significantly alters how LLM interprets the real world. While other cameras capture raw image "snapshots," the HUSKYLENS 2 provides an organized, intelligent brief. It can notify a Large Language Model (LLM) that it observes a pre-identified individual performing a specified action. This provides LLMs with precise, context-aware perception, allowing them to make better, more educated decisions.
Deployment of Custom Models Trained with YOLO
Aside from the built-in models, YOLO allows users to train and deploy their own vision models on HUSKYLENS 2. HUSKYLENS 2 includes a comprehensive collection of annotation, training, and conversion tools, making it simple to build custom-trained models.
Flexible Model Combination for Diverse Applications
The HUSKYLENS 2 design allows for serial or parallel execution of many models. This capability facilitates the development of more sophisticated and subtle applications. For example, integrating hand keypoint detection and object tracking can result in complex contactless control systems in which a gesture can trigger the tracking of a specific object. This multimodel method enables layered reasoning and more intelligent interacting behaviors.
Real-time Video Streaming for Monitoring and Debugging
HUSKYLENS 2's real-time video streaming feature enables seamless data collecting and visual display. It allows for both wired (USB-C) and wireless (via an optional Wi-Fi module) image transmission. This is crucial for applications that require remote monitoring, for gathering datasets to train new models, and for debugging visual applications by providing a live display of what the sensor detects.
Replaceable Camera Module for Any Application
HUSKYLENS 2 has a modular camera module that allows it to quickly adapt to a variety of settings, including microscopic examination and long-range surveillance. Its unparalleled versatility allows for interchangeable lenses, including manual focus, microscope, and night vision modules.
Broad Compatibility with Mainstream Controllers
With standard UART and I2C interfaces, HUSKYLENS 2 connects to a wide range of microcontrollers and single-board computers, including Arduino, micro:bit, Raspberry Pi, ESP32, UNIHIKER K10, and UNIHIKER M10.
Model Hub for Ready-to-Use Vertical Models
To speed up development, a Model Hub will be established. This online platform will have a carefully chosen collection of official and user-generated models geared to certain vertical areas such as agricultural, retail, and safety monitoring. Users will be able to explore, download, and install these customized models immediately on their HUSKYLENS 2, giving them immediate access to powerful, specific AI capabilities.
Applications
- Educational robotics
- Interactive systems
- Intelligent device prototyping
Package Includes:
- HUSKYLENS 2 AI Vision Sensorx1
- M3 Screws x6
- M3 Nuts x6
- Mounting Bracket x1
- Heightening Bracket x1
- Gravity-4P Sensor Connector Cable (30cm) x1
- Dual-Plug PH2.0-4P Silicone Cable (20cm) x1
- Power Adapter Board x1
