ROHM has developed AI-equipped MCUs (AI MCUs) – ML63Q253x-NNNxx / ML63Q255x-NNNxx – that enable fault prediction and degradation forecasting using sensing data in a wide range of devices, including industrial equipment such as motors. These MCUs are the industry’s first* to independently execute both learning and inference without relying on a network connection.

As the need for efficient operation of equipment and machinery continues to grow, early failure detection and enhanced maintenance efficiency have become key challenges. Equipment manufacturers are seeking solutions that allow real-time monitoring of operational status while avoiding the drawbacks of network latency and security risks. Standard AI processing models, however, typically depend on network connectivity and high-performance CPUs, which can be costly and difficult to install.

In response, ROHM has developed groundbreaking AI MCUs that enable standalone AI learning and inference directly on the device. These network-independent solutions support early anomaly detection before equipment failure – contributing to a more stable, efficient system operation by reducing maintenance costs and the risk of line stoppages.

The new products adopt a simple 3-layer neural network algorithm to implement ROHM’s proprietary on-device AI solution “Solist-AI™.” This enables the MCUs to perform learning and inference independently, without the need for cloud or network connectivity.

AI processing models are generally classified into three types: cloud-based, edge, and endpoint AI. Cloud-based AI performs both training and inference in the cloud, while edge AI utilizes a combination of cloud and on-site systems － such as factory equipment and PLCs － connected via a network. Typical endpoint AI conducts training in the cloud and performs inference on local devices, so network connection is still required. Furthermore, these models typically perform inference via software, necessitating the use of GPUs or high-performance CPUs.

In contrast, ROHM’s AI MCUs, although categorized as endpoint AI, can independently carry out both learning and inference through on-device learning, allowing for flexible adaptation to different installation environments and unit-to-unit variations, even within the same equipment model. Equipped with ROHM’s proprietary AI accelerator “AxlCORE-ODL,” these MCUs deliver approximately 1,000 times faster AI processing compared to ROHM's conventional software-based MCUs (theoretical value at 12MHz operation), enabling real-time detection and numerical output of anomalies that “deviate from the norm”. In addition, high-speed learning (on-site) at the point of installation is possible, making them ideal for retrofitting into existing equipment.

These AI MCUs feature a 32-bit Arm® Cortex®-M0+ core, CAN FD controller, 3-phase motor control PWM, and dual A/D converters, achieving a low power consumption of approximately 40mW. As such, they are ideally suited for fault prediction and anomaly detection in industrial equipment, residential facilities, and home appliances.

The lineup will consist of 16 products in different memory sizes, package types, pin counts, and packaging specifications. Mass production of 8 models in the TQFP package began sequentially in February 2025. Among these, two models with 256KB of Code Flash memory and taping packaging are available for purchase, along with an MCU evaluation board, through online distributors.

ROHM has released an AI simulation tool (Solist-AI™ Sim) on its website that allows users to evaluate the effectiveness of learning and inference prior to deploying the AI MCU. The data generated by this tool can also serve as training data for the actual AI MCU, supporting pre-implementation validation and improving inference accuracy.

To facilitate adoption, ROHM has built an ecosystem in collaboration with partner companies, offering comprehensive support for model development and integration. Going forward, ROHM will continue to expand this ecosystem, providing more user-friendly environments by assisting with training data creation and proposing optimal implementation methods.

• Learn more about the AI MCUs (2.3MB)

About Solist-AI™ (Solution with On-device Learning IC for STandalone-AI)

Solist-AI™ is ROHM's on-device AI solution designed for edge computing applications. Drawing inspiration from the musical term "Solist (Soloist)", which signifies solo performance, this innovative solution enables real-time learning and inference directly on standalone edge devices without relying on cloud servers. Powered by ROHM’s proprietary on-device learning AI technology, Solist-AI™ is characterized by its compact design and low power consumption, contributing to the expansion of sustainable AI innovation.

・ Solist-AI™ is a trademark or registered trademark of ROHM Co., Ltd.

Product Lineup

These AI MCUs integrate a 32-bit Arm® Cortex®-M0+ core (Maximum operating frequency: 48MHz) and ROHM’s proprietary AI accelerator AxlCORE-ODL that performs learning and inference using a 3-layer neural network. On top, leveraging versatile timer functions such as 3-phase motor control PWM along with a wide range of serial interfaces like CAN FD and 12-bit A/D converter enables flexible support for control and data processing in industrial equipment, residential facilities, and home appliances.

Part No.	Data Sheet	Memory			Operating Voltage [V]	Operating Temp. Ta [°C]	Timer	No. of Serial I/F				A/D Converter	Package [mm]	Packaging Specifications	Online Sales
		Code Flash [KB]	Data Flash [KB]	RAM [KB]				CAN FD	I2C	SPI	UART
ML63Q2534-NNNTBZWAY		128	8	16	2.3 to 5.5	-40 to +105	16-bit timer (independent operation) × 6, 16-bit timer (timer/PWM/capture modes) × 2, 3-phase motor control PWM × 3 (3-phase × positive/ negative = 6 outputs), Watchdog timer × 1, Real-time clock × 1, Time base counter × 2	1	1	2	4	12bit SA-ADC: 12ch 2 units (Max. 1Msps)	TQFP48 (9.0×9.0×1.2)	Tray	–
ML63Q2534-NNNTBZWBY														Taping	In preparation
ML63Q2537-NNNTBZWAY		256												Tray	–
ML63Q2537-NNNTBZWBY														Taping	✓
ML63Q2554-NNNTBZWAY		128											TQFP64 (12.0×12.0×1.2)	Tray	–
ML63Q2554-NNNTBZWBY														Taping	In preparation
ML63Q2557-NNNTBZWAY		256												Tray	–
ML63Q2557-NNNTBZWBY														Taping	✓
☆ML63Q2534-NNNGDZW5AY		128											WQFN48 (7.0×7.0×0.8)	Tray	–
☆ML63Q2534-NNNGDZW5BY														Taping	–
☆ML63Q2537-NNNGDZW5AY		256												Tray	–
☆ML63Q2537-NNNGDZW5BY														Taping	–
☆ML63Q2554-NNNGDZW5AY		128											WQFN64 (9.0×9.0×0.8)	Tray	–
☆ML63Q2554-NNNGDZW5BY														Taping	–
☆ML63Q2557-NNNGDZW5AY		256												Tray	–
☆ML63Q2557-NNNGDZW5BY														Taping	–

☆: Under Development

AI MCU Development Support Tools

ROHM AI MCUs utilize a standard Arm® core, ensuring compatibility with commercially available tools as well as ROHM’s proprietary integrated development environment. To evaluate learning and inference, an AI operation verification simulator is provided, along with a real-time viewer for assessing AI effectiveness.
Further details on the AI MCU development support system and an overview of each product can be found on ROHM’s dedicated AI MCU development system support page (below).
https://www.rohm.com/lapis-tech/product/micon/solistai-software

■ ROHM Website Resources
　Solist-AI™ Sim: PC-executable simulator for verifying AI operation
　Solist-AI™ Scope: Real-time viewer for assessing AI effectiveness (included with reference software)
　Reference Software: Sample software for AI MCUs
　Integrated Development Environment: LEXIDE-Ω (developed by ROHM)

■ Available Products
　Arm® Integrated Development Environment: Arm® Keil® MDK
　Arm® Debug Adapter: Debugger for connecting a computer to the Arm® core
　USB-SPI Conversion Adapter: Adapter for connecting the AI MCU to Solist-AI™ Scope

■ Online Sales
　MCU Evaluation Board: Board for standalone AI MCU evaluation/software development

Online Sales Information

• AI MCU Product Information
　Sales Part Nos: ML63Q2537-NNNTBZWBY、ML63Q2557-NNNTBZWBY
• MCU Evaluation Board Information
　Sales Part Nos: RB-D63Q2537TB48、RB-D63Q2557TB64

Application Examples

Factory Automation (FA) sensors, motors, batteries, power tools, residential facilities, home appliances, robots. Other uses include devices requiring fault prediction, equipment where operational downtime is unacceptable, and systems that demand improved prediction accuracy.

AI MCU Use Cases

High-precision anomaly detection and condition monitoring is possible by combining the AI MCU with various sensors.
■ FA Sensor + AI MCU
　Utilizes data such as light, temperature, flow rate, and sound to monitor equipment conditions and detect anomalies. Also performs anomaly detection and degradation forecasting of the sensing units themselves, including FA sensors.
■ Motor + AI MCU
　Monitors motor current, temperature, and rotational speed to detect load abnormalities and bearing damage.
■ Accelerometer + AI MCU
　Tracks vibration levels to implement condition-based maintenance (CBM) using criteria tailored to each specific machine.
■ AE (Acoustic Emission) Sensor + AI MCU
　Allows for the ultra-early detection of mechanical anomalies by collectively analyzing key AE indicators (peak amplitude, average value, energy, and count).
■ Residential Facilities/Home Appliances + AI MCU
　Leverages data from existing sensors to detect equipment anomalies at an early stage, determine maintenance requirements, estimate non-sensed parameters, and predict the time needed for specific operations.
■ Industrial Robots + AI MCU
　Detects anomalies and the optimal adjustment timing of various robot components using edge (endpoint) sensors and the AI MCU, transmitting only the results to the main CPU.

Terminology

3-Layer Neural Network

A neural network (a model of mathematical formulas and functions) inspired by the human brain with a processing simple three layers comprised of input, intermediate, and output layers. Deep learning involves adding dozens of intermediate layers to achieve more complex AI processing.

PLC (Programmable Logic Controller)

A programmable control device used for the automated control of production equipment and industrial machinery. Equipped with an MCU, PLCs are widely employed for automation and machine control, where high environmental durability and reliability are essential.

AI Accelerator

Specialized hardware designed to improve processing speed by offloading AI tasks from software-based CPU execution to efficient hardware-level processing.

CAN FD (CAN with Flexible Data Rate)

An enhanced communication protocol that extends the conventional CAN (Controller Area Network) specification to achieve faster data transfer speeds and larger data payloads. While traditionally used in automotive networks, CAN FD is now increasingly adopted in the industrial sector, supporting real-time control applications such as PLCs, robotic systems, and motion control equipment.

3-Phase Motor Control PWM (Pulse Width Modulation)

A function that generates PWM signals to efficiently control the operation of 3-phase motors. Allows precise control of the motor’s rotational speed and torque by outputting pulse signals at three different timings. This function is essential for delivering energy-efficient, high-performance motor drive in industrial equipment and home appliances, enabling real-time control through the built-in MCU.

A/D Converter

A function that transforms analog signals (i.e. continuous data like voltage or current) into digital (numerical data). It converts sensor inputs like temperature, vibration, and pressure, into a format the MCU can process. This capability plays a vital role in supporting real-time processing in sensor data acquisition and control systems.

Condition-Based Maintenance (CBM)

A method that involves real-time monitoring of machinery and equipment to perform timely maintenance when anomalies are detected. Utilizing an AI MCU makes it possible to learn the unique characteristics and installation environments of each machine, enabling highly tailored and efficient CBM for individual equipment.

AE (Acoustic Emission) Sensor

A sensor that detects subtle high-frequency sounds (elastic waves) generated within materials to monitor for damage or anomalies in structures and equipment. When combined with an AI MCU, real-time learning and inference of early-stage anomalies are possible, enabling predictive fault detection.

Note: DigiKey™, Mouser™ and Farnell™ are trademarks or registered trademarks of their respective companies.
Arm®, Cortex®, and Keil® are registered trademarks of Arm Limited (or its subsidiaries) in the US and other countries.