Power-efficient Edge AI processing targets next-gen HMI

Power-efficient Edge AI processing targets next-gen HMI

Members box download this article in PDF format.

We’ve been steeped in traditional human-machine interface (HMI) methods for so long that adding artificial intelligence (AI) and edge computing necessitates a fine-tuning of our knowledge base. In the past, an HMI consisted of a physical control panel with push buttons, switches, trackballs, and LEDs that allowed users to communicate with a machine. But it is no sacrilege to suggest that these quickly leave the scene.

Instead, edge computing brings computing power to where your data is collected. Edge AI is more nimble and nimble than other forms of data processing, where data is sent to remote data centers or the cloud for processing. It offers faster localized processing with less latency than traditional forms of cloud computing.

With the proliferation of AI and machine learning (ML), the next generation of HMIs is poised to bring whole new ways to interact with machines and devices. Coupling advanced AI features such as machine vision, analytics, and predictive maintenance with HMI applications is changing the way we interact with machines. Examples include gesture and/or facial recognition in a noisy factory environment or enabling control of machinery via digital voice assistants (such as Alexa and Siri) and wireless connection.

Sponsored Resources:

Taking HMI to the next level

As HMI continues to evolve, the processor technology behind these applications must be ready to support this evolution. Recently, Texas Instruments launched new state-of-the-art artificial intelligence processors for HMI applications. Designed for low power consumption, the first devices in TI’s Sitara AM62 processor family, the AM625 and AM623, bring state-of-the-art, power-efficient AI processing to dual-display and small-form-factor applications. They include an AI accelerator that achieves high performance of deep learning AI inference. Inference involves the process of AI analyzing data collected from a sensor and then making decisions based on it in real time or near real time.

TI has put a lot of effort into the development of the AM62x family (Fig.1). The processors consist of a 64-bit quad-core Arm Cortex-A53 microprocessor, a single-core Cortex-R5F microcontroller (MCU), and a Cortex-M4F microcontroller. While the AM623 targets IoT and gateway SoCs with Arm Cortex-A53-based object and gesture recognition, the AM625 is specifically aimed at human-machine interactions with state-of-the-art Arm Cortex-based AI -A53 and Full HD dual display capability. .

The best part may be the low power consumption of these devices. AM62x processors achieve power consumption of less than 500 mW for a single A53 core running at 1 GHz. That’s nearly half of what similar low-power, low-cost processors of previous generations achieved. AM62x CPU power may be less than 1W when A53 quad cores running stressful application on all quad cores at 1.4GHz.

To bring analytics to low-power edge devices, including suspend states as low as 7 mW without the need to design for thermal considerations, TI engineers employed a few clever techniques.

Low power design

Low power consumption has been made possible by a simplified power supply architecture: the device has only two dedicated power rails and five power modes. The modular architecture of this device delivers performance with support for multiple low-power modes without sacrificing critical system resources such as connectivity, power, security, safety, and cost.

Deep sleep mode at <5mW enables longer battery life, while active power of <1.5W is enabled by a core voltage of 0.75V. As a result, processors AM62 can reduce power consumption in industrial applications by up to 50% compared to competitive devices, allowing an application powered by AA batteries to stay on for over 1,000 hours.

Achieving optimal power performance is made even easier with the new TPS65219, a power management integrated circuit (PMIC) specifically designed to meet the power requirements of the AM62x processor. The device is characterized over an ambient temperature range of –40 to +105°C, making it suitable for various industrial applications. It includes three synchronous buck DC-DC converters and four linear regulators.

Additional Features

AM62x processors include on-chip resources such as universal asynchronous receiver/transmitter, multi-channel SPI, and I2C, and various connection options for common industrial sensors or controllers. These processors also offer dual Ethernet support and EtherCAT master support with a third-party ecosystem.

The parts support a wide variety of display interfaces, including the RGB888 color format, where 8 bits are used for each of the red, green, and blue color components, and a low-voltage differential signaling interface that takes support 2K and high definition displays. The processors support dual displays, providing engineers with design flexibility as dual displays effectively expand the visual space and help make workflow smoother.

Additionally, AM62x processors support multiple operating systems, including major Linux and Android operating systems.

The set of peripherals included in AM62x enables system-level connectivity such as USB, MMC/SD cards, camera interface, CAN-FD and GPMC (general purpose memory controller) for parallel host interface to an ASIC /external FPGA. The AM62x device also supports secure boot for IP protection with an integrated hardware security module (HSM) and advanced power management support for portable and power-sensitive applications.

The AM625 and AM623 processors come in a 13×13mm, 425-pin ALW package, with pricing starting at less than $5 per 1000 unit quantity. The processors can meet the AEC-Q100 automotive standard in the 17.2 × 17.2 mm (AMC) package. Industrial and automotive functional safety requirements are met when using the integrated Cortex-M4F cores and dedicated peripherals, all of which can be isolated from the rest of the AM62x processor.


A rich hardware ecosystem, including a third-party evaluation module (EVM), helps designers get up and running on application designs faster. The SK-AM62 starter kit for advanced Arm-based industrial HMI, IoT and AI is a low-cost standalone test and development platform ($149) (Fig.2). It offers power-efficient processing (with Cortex-A53, -R5F, and -M4F cores), Full HD and dual-display support (HDMI and LVDO panels), and software that includes demos for cutting-edge AI , HMI, etc.


The AM62x’s low-power modes and low active power consumption enable a wide range of battery-powered uses, and its small form factor makes it easy to design products without heatsinks or fans. Additionally, the 0.75V base voltage power management and operation features allow performance and power to be tuned for each application, resulting in a simple and low-cost power solution. expensive.

Sponsored Resources:

#Powerefficient #Edge #processing #targets #nextgen #HMI

Leave a Comment

Your email address will not be published.