Qualcomm® Robotics RB6 Platform

Delivering power-efficient, advanced edge-AI computing and video processing performance with 5G connectivity for autonomous robotics.

The Qualcomm Robotics RB6 Platform powers the next-generation robotics and intelligent machines including autonomous mobile robots (AMRs), delivery robots, highly automated manufacturing robots, urban air mobility (UAM) aircrafts, autonomous defense solutions, and beyond.

The Qualcomm Robotics RB6 platform sits at the premium tier of Qualcomm Technologies’ robotics solutions, delivering expanded capabilities that take enterprise and industrial robotics innovation to the next level with enhanced AI and 5G. The new solution delivers state-of-the-art 5G connectivity with support for global sub-6GHz and mmWave bands in mainstream, enterprise, and private networks.

Qualcomm Technologies robotics platforms have flexible architecture with expansion cards to support evolving connectivity features allowing Qualcomm Robotics RB6 platform to deliver support for 3GPP Release 15, and Releases 16, 17 and 18 features as cards become available in the future. The elevated platform brings best-in-class, premium, edge AI and video processing capabilities through the enhanced Qualcomm AI Engine, with support for 70-200 Trillion Operations Per Second (TOPS).

Applications:

Qualcomm® Robotics RB6 Platform

Purpose-built for the future of robotics

Robotics RB6 Platform Applications

Delivery Robots

Autonomous mobile robots (AMRs)

Urban air mobility (UAM) transportation

Automated manufacturing robots

Collaborative robots

Robotics RB6 Platform Highlights

All-in-one hardware solution

Your camera, computer vision, AI and 5G hardware development needs are integrated into one platform

Easy-to-use, comprehensive, customizable premium AI SDK

The Qualcomm® Intelligent Multimedia SDK brings together multimedia, AI and ML, computer vision (CV), and networking building blocks to support end to end deployment of robotic applications.

Enhanced Qualcomm® AI Engine and power-efficient on-device AI computing and video processing

The enhanced Qualcomm AI Engine provides 70-200 TOPS (INT8) of performance at very low power and delivers maximum performance per watt. Also our powerful image signal processor (ISP) provides support for up to seven concurrent cameras, or up to 24 simultaneous video streaming cameras.

Latest 5G connectivity

Qualcomm Robotics RB6 platform provides 5G connectivity with support for global sub-6GHz and mmWave bands in mainstream, enterprise, and private networks. The platform’s flexible architecture with expansion cards supports evolving connectivity features allowing Qualcomm Robotics RB6 platform to deliver support for 3GPP Release 15, and Releases 16, 17 and 18 features as cards become available in the future.

Robotics RB6 Platform Features

• Qualcomm® Spectra™ 480 Image Signal Processor designed to deliver a premium camera experience that can process 2 Gigapixels per second with high-performance capture of 200-megapixel photos, 8K video recording and 4K HDR video capture.

• Qualcomm® Adreno™ 650 Visual Processing Subsystem deliver’s quality graphics for larger than-life immersive experiences using the Adreno graphics processing unit (GPU) and video processing unit (VPU).

• Enhanced Qualcomm AI® Engine with Qualcomm Hexagon DSP for on-device AI processing at the edge and delivers premium computer vision (CV) experiences for high-performance use cases.

• Qualcomm® Kryo™ 585 CPU: Manufactured in 7nm process node, optimized across four high-performance Kryo Gold cores and four low power Kryo Silver cores.

• Qualcomm® Secure Processing Unit (SPU) offers vault-like security that is designed to help safeguard your facial data, iris scan and other biometric data. It supports hardware root of trust, Qualcomm® Trusted Execution Environment, Secure boot and camera security.

Compatible With

Specifications of the Qualcomm® Robotics RB6 Development Kit

Component Qualcomm® Robotics RB6 Development Kit
CPU CPU Cores:  Qualcomm® Kryo™ 585 CPU
CPU Clock Speed: Up to 2.84 GHz
GPU GPU Name: Qualcomm® Adreno™ 650 GPU
API Support: OpenCL™, OpenGL® ES
DSP DSP Technology: Qualcomm® Hexagon™ Tensor Accelerator
Qualcomm® Artificial Intelligence (AI) Engine Hexagon Processor: Qualcomm® Hexagon™ Scalar Accelerator
Memory Density: Up to 16 GB
Wi-Fi Standards: Wi-Fi 6-ready, 802.11ad, 802.11ay, 802.11ac Wave 2, 802.11a/b/g/n
Security Support Wi-Fi Security: WPA3-Enterprise, WPA3-Enhanced Open, WPA3 Easy Connect, WPA3-Personal, FIPS 140-2 support
Security Features: Camera Security, Crypto Engine, Cryptographic Accelerator, Qualcomm® Trusted Execution Environment (TEE), Secure Boot
Camera Video Capture Features: 8K video capture at 30fps, Up to 10-bit color depth video capture, 4K video capture with simultaneous 64MP photo capture, 4K Video Capture @ 120 FPS
Video Capture (30 FPS): 4K HDR video capture
Single Camera: Up to 200 MP
Dual Camera, ZSL, 30fps: Up to 25 MP
Single Camera, ZSL, 30fps: Up to 64 MP
Data Types Data Types: INT8, INT16, FP16, FP32
Part Part Number(s): QRB5165N, Cloud AI 100
Card Card Types: Dual M.2, Dual M.2 (edge)

for more information, please visit:Qualcomm® Robotics RB6 Platform

Get Support

Help Center

Technical Support

(for cusmtomer)

Qualcomm® Robotics RB6 Development Kit (Standard 70 Tops)
Standard version: without expansion 5G Mezz Board.
Package List: C5165N SOM &Interposer+Main IO+Nav Mezz+AI Mezz+AIC module (70 Tops).

$2199

Qualcomm® Robotics RB6 Development Kit (Standard 200 Tops)
Standard version: without expansion 5G Mezz Board.
Package List: C5165N SOM &Interposer+Main IO+Nav Mezz+AI Mezz+AIC module (200 Tops).

Qualcomm® Robotics RB6 Development Kit:

5G Mezz package(Sub-6G only, Japan&Korea), Thundercomm T55M
With 5G mezzanine board and Thundercomm 5G NR module T55M-EA, offers the 5G NR Sub-6GHz connectivity in Asia on core kit or vision kit.
5G Mezz package(Sub-6G only, North America&Europe), RM502Q-AE
with 5G mezzanine board and 5G NR module RM502Q-AE, offers the 5G NR Sub-6GHz connectivity in North America and Europe on core kit or vision kit.

Accessories:

Everything You Need to Generate System Image

 

RB6 SDK Manager provides an end-to-end image generation/downloading solution for developers to work with RB6 devices.


Highlights

• Generate system image on local PC

• Version control of official image releases

• Flash full build into RB5 device

• Designed for developers

System Requirements

• Host OS: Ubuntu Desktop 18.04

• Alternative: Ubuntu 18.04 on Docker

NOTE: Ubuntu under libvirt KVM/QEMU is not supported

What’s in RB6 SDK Manager

• SDK Manager debian package

• README (User guide)

• Dockerfile for generating Ubuntu 18.04 docker image


Release Note for Updated Image:

Release note for LU image
QRB5165.UBUN.1.1-220715:
1. Based on Qualcomm release r00023.1a
2. Integrated Cloud AI (AIC) Platform Software Development Kit (SDK) [Version 00060.1]
3. Debug mode with debug logs enabled
Notes: please refer to Qualcomm official release notes for complete support lists of QC releases

Release Note for SDK Manager

New Features in SDK manager v3.0.0:

-Support for RB6 Non-pop

New Features in SDK manager v2.1.0:

-Enable LE SDK toolchain download

New Features in SDK manager v2.0.0:

- Commands to download LE releases

New Features in SDK manager v1.2.0:

- Update README

New Features in SDK manager v1.1.0:

- Modify flash method

- Update README

New Features in SDK Manager v1.0.1:

- Update README

- Add debug message on flashing

- Minor fixes

New Features in SDK Manager v1.0.0:

- User credential check

- Dynamic image version control

- Enable system image repack

- Enable full build flash on Ubuntu


Troubleshoot

1. Internet Timeout Issue

Internet timeout issue may happen during the image generation process. If issues like "Unable to fetch" are encountered, try to run command 1 again.

2. Apt Source issue

If the download fails, check the internet connection and the source list.

3. Device Boots Up

The device reboots after the flashing process is completed. If the sdkmanager does not detect the device, execute the followings:

1) If Ubuntu 18.04 is used on Docker, check whether "adb kill-server" is entered on the host PC before flashing image.

2) Reboot the device manually, open a new terminal window and enter 'adb shell' to check device.

3) Check if any debian packages are modified.

4. If flash process on Ubuntu systems does not work properly, copy full-build folder to a Windows PC and use Thundercomm MULTIDL_TOOL to flash the image. Refer to Here for the tool and user guide.

5. SDK manager flash issue

If issue like 'qdl failed, sahara failed' is encountered, enter the following command on host machine before restart the flash.