Chris Catterton, Director of Solution Engineering, ONE Tech
Chris Catterton is the Director of Solution Engineering at ONE Tech. Located in Dallas, Texas, Catterton has over 10 years of experience deploying global enterprise software solutions- primarily in the space of Industrial IoT, Process Automation and Embedded AI.
Bob Graham, Senior Engineer, ONE Tech
Bob Graham is a Senior Engineer at ONE Tech. Located in Dallas, Texas, Bob has over 20 years of experience in the hardware design space focused on embedded controls systems.
Latency and bandwidth challenges associated with using cloud services can lower Microcontroller Units’ (MCUs) efficiency in a broad range of industrial, automotive, medical, consumer, and many other embedded applications. Machine learning (ML), on the other hand, allows systems to learn process patterns, adapt to changing situations, and make intelligent decisions in real-time. In modern automobiles, AI-enabled systems and electronic sensors enable several new functions for improving the safety and driveability of vehicles, such as Advanced Driver-Assistance -Systems (ADAS), smart automotive manufacturing, and driverless vehicles. ML also provides greater operational awareness to operators and managers for predictive maintenance and optimizing production, and minimizing wastage in industrial settings. AI models can be trained on many of today's high-performance CPUs or GPUs and implemented on MCUs per system requirements.
AI-enabled MCUs implement a host of functions for enhancing the ease of use and efficiency of machines or devices, including facial detection and recognition, voice control, and Natural Language Processing (NLP). There are three main methods for implementing AI functions in edge and node designs; using pre-trained neural networks optimized for MCUs hosted on the Cloud, integrating AI libraries (with associated AI-training codes) into MCUs, and using AI-dedicated co-processors. A key advantage of AI-enabled MCUs is that they implement ML functions at lower costs with ultra-low power consumption.
Real-time responsiveness implemented through machine learning makes MCUs less vulnerable to cyber-attacks. Renesas' MCUs powerful security features for next-gen embedded applications, such as the following product families:
The first-generation RA devices incorporate a selection of hardware-based security features, from simple AES acceleration to fully integrated crypto subsystems (Secure Crypto Engines or SCEs) isolated within the MCU. The RA6 Series also offers the added benefit of NIST CAVP certification plus PSA Level 1 and PSA Level 2 Certifications. Future RA devices will further expand the MCU security capabilities with the addition of next-gen SCEs, Arm®v8 M TrustZone®, and Trusted Firmware M, validated by additional NIST and PSA certifications.
The Renesas Synergy MCUs, supported by the Synergy Software Package, provide hardware security features previously available only in dedicated Secure Element chips developed for specialized, highly secure identity applications (e.g., SIM cards, VPN tokens, etc.) The MCU's high integration level eliminates the need for another chip, a socket, and a communications bus (I2C bus). This integration level reduces the size, complexity, power consumption, and cost of the circuit board.
Renesas RX security solutions implement a Root of Trust for IoT devices using encryption by key data protected by a strong Trusted Secure IP and a memory-protected authentication program. By implementing security functions using an RX microcontroller (MCU), users can better protect their IoT devices against threats.
In this webinar, you’ll learn how MicroAI™ is revolutionizing the industry with next-gen embedded AI. ONE Tech's MicroAI™ is a self-correcting, semi-supervised learning engine that aggregates data from PLCs, IoT sensors, and other existing sources. This engine tunes itself to create a 360-degree view of asset behavior, delivering significant performance improvements and security enhancements to any IoT device or machine. By bringing AI training to the MCU, OEMs can significantly reduce the amount of data that leaves the asset. MicroAI™ enables automated machine learning model formation for anomaly detection without relying on labeled data. This results in significantly faster deployments of AI-powered machine learning.
by Omron
