Collaboration between Arrow and ST in the field of AMR

The collaborative engineering initiative addresses the integration challenges faced by original equipment manufacturers when combining real-time motion control with high-level artificial intelligence processing. By combining a dedicated real-time control architecture with advanced computational modules, the platform simplifies the development of autonomous mobile robots for deployment in logistics centers, manufacturing facilities, and automated laboratories. The system coordinates sensor fusion, localization, and power regulation within a unified footprint, minimizing the engineering risks typically associated with custom subsystem validation.

Technical Architecture and Sensor Fusion Mechanisms
The core architecture utilizes an NVIDIA Jetson Orin Nano compute module running an industrial Robot Operating System 2 stack to handle high-level processing tasks, including simultaneous localization and mapping. Real-time motor actuation and low-level sensor data aggregation are managed by a dedicated STM32 microcontroller board.

Motion control is executed via dual brushless direct current motor drives utilizing STSPIN32 and STDRIVE controllers, which regulate wheel velocities based on feedback from integrated micro-electromechanical systems inertial measurement units, magnetometers, and environmental sensors. The primary navigation pipeline ingests data from external lidar and vision sensors, processing the inputs through standard Nav2 and Cartographer algorithms to achieve dynamic obstacle avoidance and precise indoor positioning.

The electrical subsystem features power and battery management circuits optimized for 24-volt operations, designed with an architectural path to accommodate 48-volt industrial power grids. This power distribution system ensures consistent current delivery to the drive motors while maintaining isolated voltage rails for the digital computing components.

Additional Context
This section details technical specifications and competitive benchmarking not included in the original product announcement.

The platform enters a market segment where development time and component interoperability are critical metrics for industrial automation. To evaluate its position, the reference design can be compared against standard custom integration approaches and existing open-source robotics hardware frameworks.

While generic industrial single-board computers require developers to independently design power distribution boards, write custom motor driver abstraction layers, and validate sensor timing synchronization, this reference design pre-integrates these elements. Utilizing specialized silicon like the STSPIN32 reduces physical board space and propagation delay compared to setups relying on discrete microcontroller and gate driver layouts connected over standard serial buses.

Edited by Evgeny Churilov, Induportals Media - Adapted by AI.

www.arrow.com