Qualcomm today launched the Robotics 5G platform, which it claims is the first in the world to support AI acceleration and 5G. It takes the high end of the chip manufacturer's pre-built robotics solutions and was developed to enable things like machine learning, heterogeneous computing and computer vision, while withstanding temperatures of "industrial quality" and including security "at every level".
Stakeholders like Qualcomm believe that AI is marginalized. TrendForce predicts The edge computing product segment will grow by an average annual growth rate of more than 30% from 2018 to 2022 rangerThe edge cloud service market will grow by at least 50% in 2020 as vendors deploy new infrastructure-as-a-service and cloud-native programming products on distributed devices.
"(W) With billions of connected (communicating devices) and the huge amount of data that is being activated, robotics is one of the beneficiaries (of 5G and AI) – it will benefit from these technologies," said Dev Singh, Senior Director of Qualcomm VentureBeat in a telephone interview. "We believe that (5G and AI) will be really, really elementary for future robotics."
Qualcomm is picking up on the trend with the RB5 platform, which builds on the RB3 platform launched early last year. Like the RB3, the RB5 includes a range of hardware, software, and development tools for use cases in the consumer, corporate, defense, industrial, and professional fields.
It includes a Qualcomm QRB5165 processor tailored for robotics applications with a heterogeneous computer architecture combined with Qualcomm's fifth generation AI engine. An image signal processor (ISP) supports up to seven simultaneous cameras to enable simultaneous localization and assignment (SLAM), object recognition and classification, autonomous navigation and path planning. In addition to Qualcomm's Hexagon Tensor Accelerator, a companion modem that works with 4G and 5G networks, and native support for the EtherCAT standard, an industrial protocol based on Ethernet with integrated security mechanisms, there is a special engine for video analysis.
"(5G) will drive the (next) industrial revolution – it will be the connectivity structure for the factory of the future," said Singh. “5G is the first wireless technology that can handle real-time latency … and brings with it things that replace many of the traditional communication protocols. 4G can do a lot, but it is not so reliable – errors can occur, e.g. B. Connection failure. 5G offers more for your money, especially for mission critical services. "
A little more about the QRB5165 of the RB5: It contains a Qualcomm Kryo 858 with octackers, an Adreno 650 GPU and discrete digital signal processors for computers, audio and sensors. The ISP – the Spectra 380 – can process 2 gigapixels per second and offers Dolby Vision video recording, 8K video recording at 30 frames per second, 200 megapixel photos, simultaneous recording of 4K HDR videos at 120 frames per second and 64MP – Photos with zero shutter lag. Another chip – Qualcomm's Secure Processing Unit (SPU) – includes secure startup, cryptographic accelerators, malware protection, a secure token for remote confirmation and provision of secure devices, and biometric authentication (including fingerprint, iris, voice) and face).
According to Qualcomm, the QRB5165, together with the fifth-generation AI engine, delivers AI performance of up to 15 TOPS (TOPs relate to tera operations per second). That is compared to the 3 TOPS of the RB3.
In developing RB5, Qualcomm entered into a strategic collaboration with TDK, the Japanese multinational electronics company that manufactures electronic materials, components, and recording and data storage media. TDK supplied the six-axis inertial measurement unit, the bottom-port microphone and the robot application algorithms in the RB5 as well as the sensors for Hall effect, temperature and air pressure.
RB5 is part of the RB5 Development Kit, a customizable platform that provides support for Linux, Ubuntu and robotics middleware such as Robot Operating System (ROS) 2.0 and pre-integrated drivers for cameras, sensors and 5G connectivity. It is compatible with OpenCL, OpenGL and OpenCV as well as with depth detection cameras such as the RealSense Depth Camera D435i from Intel and the TOF camera from Panasonic. And it is available in a variety of flavors that package TDK's ultrasonic time-of-flight and embedded motor control solutions for commercial and industrial temperature ranges.
Commercial products based on the RB5 platform are expected to arrive in 2020. However, a development kit is now available from Thundercomm. (Customers can choose extended lifecycle support until 2029.) According to Qualcomm, over 20 early adopters are evaluating the platform, and more than 30 of the company's ecosystem partners are developing support hardware and software to enable robotics applications. Partners include AirMap, Fractal.ai, Canonical and ModalAI.
“The most important things for a robot – the key elements of a robot – are recognizing, thinking, acting and communicating. This platform brings them all together, ”said Singh, noting that Qualcomm's robotics offerings are already driving millions of robots from companies like iRobot, Sony and LG. "We all know that robotics will be an integral part of (segments like) autonomous provision and autonomous management of business-critical activities. We are very, very excited about it and double the market. "