ICs & Semiconductors Archives - Page 167 of 235

AI and robotics starter kit

Mon, Oct 12, 2020

A developer kit to teach AI has been created by Nvidia. The Nvidia Jetson Nano 2Gbyte developer kit is aimed at a new generation of students, educators and hobbyists, Nvidia said at last week’s GTC.

The Jetson Nano 2Gbyte developer kit is supported by Nvidia’s free online training and AI-certification programs, to supplement the many open-source projects, how-tos and videos contributed by thousands of developers in the Jetson community.

The kit is the latest offering in Nvidia’s Jetson AI at the Edge platform, which ranges from entry-level AI devices to advanced platforms for fully autonomous machines.

It is supported by the company’s JetPack software development kit (SDK), which comes with Nvidia container runtime and a full Linux software development environment. This allows developers to package their applications for Jetson with all its dependencies into a single container that is designed to work in any deployment. And it is powered by the same CUDA-X accelerated computing stack used to create breakthrough AI products in such fields as self-driving cars, industrial IoT, healthcare and smart cities.

The Jetson Nano 2Gbyte developer kit can run a diverse set of AI models and frameworks and provides a scalable platform for learning and creating AI applications as they evolve.

It has been endorsed by organisations, enterprises, educators and partners in the embedded computing ecosystem. For example, Jim McGregor, principal analyst at Tarias Research, said: “Nvidia’s Jetson is driving the biggest revolution in industrial AIoT. With the new Jetson Nano 2GB, NVIDIA opens up AI learning and development to a broader audience, using the same software stack as its data center AI computing platform.”

Matthew Tarascio, vice president of Artificial Intelligence at Lockheed Martin, said: “Acquiring new technical skills with a hands-on approach to AI learning becomes critical as AIoT drives the demand for interconnected devices and increasingly complex industrial applications. We’ve used the Nvidia Jetson platform as part of our ongoing efforts to train and prepare our global workforce for the AI revolution.”

Emilio Frazzoli, professor of Dynamic Systems and Control at ETH Zurich, said: “The Duckietown educational platform provides a hands-on, scaled down, accessible version of real-world autonomous systems. Integrating Nvidia’s Jetson Nano power in Duckietown enables unprecedented, affordable access to state-of-the-art compute solutions for learning autonomy.”

It has also been used as part of the STEM curriculum at Boys & Girls Club of Western Pennsylvania, USA.

The Jetson Nano 2GB Developer Kit will be available at the end of the month for $59 through NVIDIA’s distribution channels.

http://www.nvidia.com

> Read More

Image sensor minimises distortion in machine vision and mixed reality applications

Fri, Oct 09, 2020

A combination of a high dynamic range and pixel design means that the AR0234CS CMOS global shutter image sensor by ON Semiconductor delivers image clarity at 120 frames per second.
The 2.3 Mpixel CMOS image sensor, with global shutter technology is designed for a variety of applications including machine vision cameras, augmented reality (AR), virtual reality (VR) and mixed reality (MR) headsets, autonomous mobile robots (AMRs) and barcode readers.

The AR0234CS captures 1080p video and single frames, up to 120 frames per second. It is claimed to have industry-leading shutter efficiency, to produce crisp and clear images by minimising frame-to-frame distortion in high speed scenes and reducing the motion artefacts that other image sensors experience.

The pixel architecture delivers high dynamic range to support lighting conditions from the darkness of night to bright sunlight. The low noise and improved low light response makes it suitable for applications spanning across consumer, commercial and industrial IoT. The extended operating temperature range makes it deployable in challenging outdoor conditions.

As manufacturers automate production, using vision-based systems, the need for quality image sensing is increasing, explains Gianluca Colli, vice president and general manager, Industrial and Consumer Sensor Division (ICSD) Group at ON Semiconductor. As a result, they are demanding optimum size, performance and power in image sensors, he continued.

The sensor also has programmable regions of interest with on-chip histogram, auto exposure control and 5 x 5 statistics engine, fully integrated strobe illumination control, a flexible row and column skip mode, along with horizontal and vertical mirroring, windowing and pixel binning.

Together with the AP1302 image signal processor (ISP), the AR0234CS delivers a comprehensive camera system that can be designed and developed quickly for fast time-to-market, says ON Semiconductor. System designers can access the DevSuite software to evaluate features and capabilities, configure and tune the sensor, and provide a ready-made output that is usable for further image processing.

The AR0234CS is offered in colour and mono variants, with 0 or 28 degree chief ray angle (CRA).

Samples and development hardware are available now through local ON Semiconductor sales support representatives and authorised distributors.

http://www.onsemi.com

> Read More

Neural network accelerator chip enables IoT AI in battery-powered devices

Thu, Oct 08, 2020

To reduce energy consumption and latency by a factor of over 100 to enable complex embedded inference decisions at the IoT edge, Maxim Integrated has developed the MAX78000 low power neural network accelerated microcontroller.

It moves AI to the edge without performance compromises in battery-powered internet of things (IoT) devices, says the company. It is able to execute AI inferences at less than 1/100th the energy of software solutions to dramatically improve the run-time for battery-powered AI applications. It also enables new, complex AI use cases.

The MAX78000 executes inferences 100x faster than software running on low power microcontrollers and at a fraction of the cost of FPGAs or GPUs, continues Maxim.

Rather than gathering data from sensors, cameras and microphones, sending that data to the cloud to execute an inference, then sending an answer back to the edge, which is challenging due to poor latency and energy performance. Using low power microcontrollers can be used to implement simple neural networks but latency suffers and only simple tasks can be run at the edge. By integrating a dedicated neural network accelerator with a pair of microcontroller cores, the MAX78000 overcomes these limitations, enabling machines to see and hear complex patterns with local, low-power AI processing that executes in real-time.

Applications such as machine vision, audio and facial recognition can be made more efficient as the MAX78000 can execute inferences at less than 1/100th the energy required by a microcontroller. The MAX78000’s specialised hardware is designed to minimise the energy consumption and latency of convolutional neural networks (CNNs). Hardware runs with minimal intervention from any microcontroller core, making operation extremely streamlined. Energy and time are only used for the mathematical operations that implement a CNN. To get data from the external world into the CNN engine efficiently, customers can use one of the two integrated microcontroller cores: the low power Arm Cortex-M4 core, or the even lower power RISC-V core.

The MAX78000EVKIT# includes audio and camera inputs, and out-of-the-box running demos for large vocabulary keyword spotting and facial recognition. Complete documentation helps engineers train networks for the MAX78000 in the tools they are used to using, either TensorFlow or PyTorch.

The MAX78000 is available from authorised distributors. The MAX78000EVKIT# evaluation kit is also available now.

http://www.maximintegrated.com

> Read More

DAC includes non-volatile memory to simplify handheld designs, says Microchip

Wed, Oct 07, 2020

Microchip claims to have overcome the problems of implementing multi-channel system control or signal outputs using DACs in portable and handheld industrial, communications, consumer or medical systems. To overcome the need for significant processor overhead for device configuration during power-up, the MCP47/48FxBx8 family of octal 12-bit DACs include non volatile memory and an integrated voltage reference (Vref) source so they can be pre-configured for safe and efficient power-up without relying on the system processor.

“Handhelds and other portable systems are expected to deliver more capabilities in smaller, simpler designs,” said Bryan J. Liddiard, vice president of Microchip’s mixed-signal and linear business unit. “We help achieve this goal with the first DACs that eliminate processor overhead during power-up and provide the channel density, low power consumption and integrated features that today’s compact systems need so they can operate over longer periods using smaller, lighter batteries,” he added.

The inclusion of non-volatile memory means that the MCP47/48FxBx8 DACs can store user-customised configuration data even when powered down. At power-up, all eight channels are then configured to the pre-defined state without burdening the system processor with this overhead.

Integrating a Vref source into the DACs reduces overall system size and complexity while providing the control to meet critical timing for safely driving all power outputs. The devices also have both SPI and I2C serial interfaces for a choice of device communication.

Operating voltage range is 1.8V to 5.5V and the DACs’ low minimum operating voltage and its power efficiency are claimed to improve thermal performance and reliability. The DACs also provide power-on/brown-out reset protection and what is claimed to be one of the industry’s fastest settling times at five micro seconds. They also operate in the extended temperature range (i.e. -40 to +125 degrees C) for industrial and automotive applications.

The MCP47/48FxBx8 family of DACs is available for volume orders. The family includes 8-, 10- and 12-bit resolution devices in a 20-lead VQFN 5.0 x 5.0mm package and a 20-lead TSSOP package.

http://www.microchip.com

> Read More

AI and robotics starter kit

Image sensor minimises distortion in machine vision and mixed reality applications

Neural network accelerator chip enables IoT AI in battery-powered devices

DAC includes non-volatile memory to simplify handheld designs, says Microchip

About Smart Cities