D-Robotics RDK X3 ROS2 AI Board (4GB, 5Tops)

The D-Robotics RDK X3 is an intelligence-native development board delivering 5 TOPS of edge AI processing alongside full ROS2 compatibility. Equipped with 4GB LPDDR4 memory and a Raspberry Pi-compatible 40-pin interface, it is purpose-built for students, educators, hobbyists, and developers building autonomous robotics, intelligent home automation, and computer vision systems.

Dedicated edge AI inference

Powered by a dual-core Bernoulli Processing Unit (BPU), the board delivers 5 TOPS of edge computing performance for fast on-device cognitive analysis. This dedicated acceleration allows real-time execution of neural networks and machine learning models for responsive robot control.

Heterogeneous multicore architecture

The board pairs a 1.5GHz quad-core ARM Cortex-A53 CPU with the dual-core BPU and 4GB of LPDDR4 RAM. This setup balances general system workloads on Ubuntu 20.04 or 22.04 with intensive AI inference tasks.

High-definition multimedia processing

Capable of handling video streams up to 4K at 60fps, the RDK X3 features hardware encoding and decoding support for H.264 and H.265 formats. Two onboard MIPI CSI interfaces allow direct integration with high-resolution camera modules for vision-based robotics.

Comprehensive peripheral connectivity

Hardware interfaces include Gigabit Ethernet, dual-band Wi-Fi, Bluetooth 4.2, an HDMI 1.4 display output, one USB 3.0 port, and two USB 2.0 ports. These connectivity options enable seamless integration with external sensors, storage, displays, and networked systems.

Raspberry Pi pinout compatibility

The 40-pin expansion header mirrors the Raspberry Pi 4B pinout, supporting GPIO, I2C, UART, SPI, I2S, and PWM. This broad compatibility allows makers and educators to easily reuse existing HATs and add-on modules.

Robotics and educational applications

Built as an all-in-one ROS2 development kit, the RDK X3 is suited for prototyping autonomous mobile robots, running higher education AI courses, and developing smart home devices.