Configured for multimodal robot tasks
An 18-core Oryon CPU, Hexagon NPU and Adreno GPU provide compute for adapted vision, language and robot policy models.
Flagship robotics development board
Rhino Pi IQ10
Bring multi-sensor perception into one compute core.
Combines an original Qualcomm IQ10 module with an APLUX robotics carrier. 64 GB LPDDR5X, 256 GB UFS, dual 10GbE and 8 GMSL2 channels bring sensor connectivity, edge inference and control peripherals onto one development platform.
350 TOPS
Maximum AI performance (INT8, dense)
64 GB + 256 GB
LPDDR5X / UFS
8 × GMSL2
Onboard camera interfaces
Dual 10GbE
Plus 1 × 2.5GbE
Bring multi-sensor perception into one compute core.
Match the compute platform to your device tasks, models, peripherals and product form factor.
High-bandwidth multisensor connectivity
Eight onboard GMSL2 channels and dual 10GbE support cameras and high-speed network sensors; expansion boards can add cameras and industrial communication.
Large memory and high-speed storage
64 GB LPDDR5X and 256 GB UFS with PCIe Gen5 x4 M.2 NVMe provide capacity and expansion bandwidth for models, vision streams and logs.
From drivers to robotics applications
A companion Ubuntu 24.04 image, board drivers and AI tools support application integration. ROS 2 and robotics references depend on the delivered software package.
From interface capabilities to device integration
Locate connectors on the hardware and plan device connections around your peripherals.
Select a number or interface name for connection details. Refer to the specification for pin assignments, voltage levels and multiplexing.
DC power
DC 10–26 V, typically 19 V. Verify connection and polarity against the specification and board markings.
Explore integration paths for real-world tasks
Plan sensor and system integration for robot perception, higher-level planning and industrial model services.
Qualcomm Dragonwing IQ10 Series
Build a robot's perception and planning hub
Up to 350 TOPS(INT8, dense)Platform-rated peak; validate performance for your workload
Embodied robot
Combine multiview perception, multimodal understanding and task planning with cameras, sensors and robot control systems.
- Platform fit
- Combine multiview perception, multimodal understanding and higher-level planning into an embodied task computing pipeline.
- Board connections
- Eight GMSL2 channels provide vision input; four CAN FD channels and high-speed networking connect robot control systems.
Multisensor inputPerception and planningAction coordination
View interfaces and integration
Low-speed autonomous driving
Develop environmental perception, localization and higher-level planning for low-speed delivery and mobile operations, integrating with chassis control systems.
- Platform fit
- Fuse environmental inputs, run localization and planning, and pass higher-level commands to the vehicle control system.
- Board connections
- Connect environmental sensors over GMSL2 and Ethernet, and the companion chassis controller over CAN FD / RS485.
Environmental perceptionLocalization and planningChassis coordination
View interfaces and integration
Industrial AI agent server
Connect industrial data, multimodal models and business tools to local compute for analytics, knowledge Q&A and task orchestration.
- Platform fit
- Combine multimodal models, industrial knowledge and business tools into local analytics and task orchestration services.
- Board connections
- Dual 10GbE + 2.5GbE connect production networks; NVMe expands model and knowledge storage. Validate workloads for the task.
Industrial data and knowledgeModels and toolsTask orchestration
View interfaces and integrationDevice images illustrate application directions; peripherals, enclosure design and software require project-specific integration.
Rhino Pi IQ10 specifications
Configurations below follow the flagship robotics development board documentation. Module capabilities and physically exposed board interfaces are distinguished.
Compute and storage
| Assembly | Original Qualcomm IQ10 module + APLUX robotics carrier |
|---|---|
| Compute platform | Qualcomm Dragonwing IQ10 Series |
| CPU | 18-core Qualcomm Oryon, up to 3.014 GHz |
| NPU | Qualcomm Hexagon |
| AI performance (INT8, dense) | Up to 350 TOPS(INT8, dense) |
| AI performance (INT8, sparse) | Up to 700 TOPS(INT8, sparse) |
| GPU | Dual Qualcomm Adreno GPUs; main GPU up to 1.0 GHz, auxiliary GPU up to 1.1 GHz |
| Memory | 64 GB LPDDR5X |
| Onboard storage | 256 GB UFS (on the carrier board) |
| Product code | A8797EZ1-152 |
Vision and display
| Camera | 8 × GMSL2, plus MIPI CSI high-speed expansion |
|---|---|
| Optional camera expansion | An expansion board adds 4 × GMSL2; not included in the base board's default interfaces |
| Display | 1 × HDMI, up to 4K / 60 Hz |
| Audio | 2 × speaker interfaces |
Connectivity and control
| Ethernet | 2 × 10GbE RJ45 + 1 × 2.5GbE RJ45 |
|---|---|
| USB-C | 3 × USB 3.1 Type-C + 1 × USB 3.1 OTG Type-C |
| USB-A | 4 × USB 2.0 Type-A |
| NVMe | 1 × M.2 NVMe，PCIe Gen5 x4 |
| Wireless expansion | 1 × M.2 Wi-Fi module interface, PCIe ×1; wireless module and antennas are optional |
| CAN | 4 × CAN FD |
| Serial ports | 2 × RS485 + 1 × RS232 |
| GPIO | 8 GPIO channels |
| High-speed expansion | PCIe Gen5 x8、Gen5 x2、MIPI CSI、USB 3.0 |
| Industrial communication expansion | EtherCAT and cellular features depend on expansion boards and delivered version |
| RTC | Onboard RTC and backup battery connector; battery supplied according to project configuration |
System and mechanical
| Operating system | Companion Ubuntu 24.04 image |
|---|---|
| Power | DC 10–26 V, typically 19 V |
| Carrier outline | 124 × 136 mm |
| Operating temperature | Board design range: -40 °C to +85 °C; complete assembly and workload require thermal validation |
| Storage temperature | Board documentation range: -45 °C to +90 °C; check accessories individually |
| Assembly | Refer to the delivered assembly drawing for height, mounting holes, cooling and peripheral clearances |
| Optional accessories | NVMe SSD, wireless module, GMSL cameras, industrial expansion boards and cables |
AI figures are platform-rated peaks. Dense, sparse and FP16 figures are listed separately; actual performance depends on the model, precision, software and cooling. Input power ratings do not represent total board consumption. Accessories and system images depend on the delivered configuration.
Product configuration and integration notes
The current board provides 64 GB LPDDR5X, 256 GB UFS on the carrier and 8 GMSL2 channels. A 12-channel configuration requires extra expansion.
Current board networking is 2 × 10GbE + 1 × 2.5GbE. Wi-Fi, NVMe SSD, cameras and expansion boards are optional according to the order.
Start development and integration with the resources
Download Chinese and English product briefs and specifications for product configuration, interface definitions and device integration requirements.
PDFPDFPDF
Product brief (Chinese)
Product overview, key capabilities and application directions.
PDF · ChineseDownload Chinese briefProduct brief (English)
Product overview, key capabilities and application directions.
PDF · EnglishDownload English briefProduct specification
Interface definitions, power requirements and mechanical dimensions.
PDF · ChineseDownload specificationRelated products
Build your edge AI product with Rhino Pi IQ10
Share your model, peripheral, mechanical and system requirements to define the product and delivery configuration together.