Qualcomm has launched its first integrated, comprehensive offering designed specifically for robotics.
This purpose-built platform features a highly optimized set of hardware, software and tools designed to help manufacturers and developers create advanced consumer, enterprise and industrial robotics products. Based on the Qualcomm SDA/SDM845 system-on-chip (SoC), the platform integrates key capabilities such as high-performance heterogeneous computing, 4G/LTE connectivity including CBRS support for private LTE networks, a Qualcomm AI Engine for on-device machine learning and computer vision, hi-fidelity sensor processing for perception, odometry for localization, mapping, and navigation, vault-like security, and Wi-Fi connectivity. The Qualcomm Robotics RB3 Platform also plans to introduce 5G connectivity support later this year.
Qualcomm’s Dev Singh, a product line manager for robotics, drones and intelligent machines, said they’d been working on this for over two years under the radar and already have more than 20 products with early customers.
Singh said robots are entering all aspects of life, from home entertainment and home care robots, to retail, warehousing and logistics as well as factory robots.
The company’s technologies are used in robotics products available today from companion robots including Anki Vector, Elli Q and Sony Aibo, to multimedia robots like Cerevo Tripon and Keecker, to labor-saving machines such as robotic vacuum cleaners from iRobot, Ecovacs and Panasonic.
The heterogeneous computing architecture powering the platform is the Qualcomm SDA845/SDM845 SoC, built on 10 nm LPP FinFET process technology. The SoC integrates an Octa Core Qualcomm Kryo CPU with performance up to 2.8GHz, a Qualcomm Adreno 630 Visual Processing Subsystem (including GPU, VPU and DPU), and a Qualcomm Hexagon 685 DSP with Hexagon Vector Extensions (HVX), delivering on-device AI processing and mobile-optimized computer vision (CV) capabilities for perception, navigation and manipulation.
The AI engine, comprising CPU, GPU, and DSP delivers up to three tera-operations-per-second (TOPS) performance, with the DSP itself delivering 1.2 TOPS at one watt with hardware acceleration. It includes the Qualcomm Neural Processing SDK, which encompasses analysis, optimization, and debugging tools designed to allow developers and manufacturers to port trained deep learning networks on the various heterogeneous compute blocks offered by the platform.
Camera and video is provided by a dual 14-bit Qualcomm Spectra 280 ISP supporting up to 32 MP single camera; support for up to 4K HDR video capture at 60 frames per second (fps). Sensor and microphone support available on the platform includes a 6-axis inertial measurement unit (IMU) consisting of a 3-axis gyroscope and a 3-axis accelerometer; capacitive barometric pressure sensor; multi-mode digital microphones; and additional ports for supplementary sensors from TDK-InvenSense.
Security is enabled with the Qualcomm Secure Processing Unit (SPU), which incorporates secure boot, cryptographic accelerators, Qualcomm Trusted Execution Environment (QTEE), and camera security. To address advanced AI, ML, and biometrics, Qualcomm SDA/SDM845 supports porting of virtualization software.
Connectivity options the platform include 4G/LTE and CBRS, with 5G support planned to be enabled late this year; Wi-Fi integrated 802.11ac 2x2 with MU-MIMO; tri-band Wi-Fi: 2.4 GHz and 5 GHz with dual band simultaneous (DBS); and Qualcomm TrueWireless Bluetooth 5.0.
The platform currently supports Linux and Robot Operating System (ROS), while also including support for the Qualcomm Neural Processing software development kit (SDK) for advanced on-device AI, the Qualcomm Computer Vision Suite, the Qualcomm Hexagon DSP SDK, and Amazon’s AWS RoboMaker, with plans for Ubuntu Linux support.
The platform’s hardware development kit contains the new purpose-built robotics-focused DragonBoard 845c development board, based on the Qualcomm SDA/SDM845 SoC and compliant with the 96 Boards open hardware specification to support a broad range of mezzanine-board expansions. Optional elements for the kit include a connectivity board; an image camera for hi-res photo, 4K video capture and AI-assisted detection and recognition of people and objects; a tracking camera for path planning and obstacle avoidance using visual simultaneous localization and mapping (vSLAM); a stereo camera for navigation; and a time-of-flight camera for people, gesture and object detection even in low light conditions.
Commercial products based on the Qualcomm Robotics RB3 platform are expected to be available in 2019. A development kit based on the Qualcomm Robotics RB3 Platform is available for purchase now starting at $499.
- CPU: Up to 2.8 GHz , Qualcomm Kryo 385 CPU 64-bit, 10 nm
- DSP: Qualcomm Hexagon 685 DSP
- Wi-Fi Standards: 802.11ac Wave 2, 802.11a/b/g, 802.11n
- Wi-Fi Spectral Bands: 2.4 GHz, 5 GHz, 60 GHz
- Channel Utilization: 20/40/80 MHz
- MIMO Configuration: 2x2 (2-stream)
- Wi-Fi Features: MU-MIMO, Dual-band simultaneous (DBS), Integrated baseband
- Bluetooth Version: Bluetooth 5.0
- Location: Satellite Systems Support: GPS, QZSS, GLONASS, SBAS, Beidou, Galileo
- Image Signal Processor: Dual 14-bit ISPs, Qualcomm Spectra™ 280 image signal processor
- Megapixel Support: Up to 16 MP dual camera, Up to 32 MP single camera
- Video Capture: Up to 4K Ultra HD video capture @ 60 FPS
- Codec Support: H.265 (HEVC), HLG, H.264 (AVC), HDR10
- Display: Standards: ULTRA HD PREMIUM-ready
- GPU Name: Qualcomm Adreno 630 GPU
- API Support: OpenGL ES 3.2, OpenCL 2.0 full
- Memory speed: 1866MHz
- Memory Type: LPDDR4x
- RAM: 8 GB RAM