A new embedded AI development platform combines high-performance processing, real-time control, camera connectivity, and expanded memory for advanced robotics and edge-computing workloads.
The D-Robotics RDK S100P (S100P) is a robotics-focused development platform designed to bring AI perception, general-purpose computing, and real-time control onto a single hardware platform. The board is positioned as an alternative for developers looking beyond conventional embedded AI platforms for demanding robotics and edge-computing applications.
The platform is built around the company’s S100P system-on-chip, which combines Arm-based application processing with a dedicated AI accelerator and a separate real-time control subsystem. This architecture allows compute-intensive AI workloads and time-sensitive control operations to run alongside each other, an approach aimed particularly at robots that need to interpret their surroundings while simultaneously responding to sensor and motor inputs.
The key features are:
- HDMI 1.4 output supporting 2560 × 1440 at 60 Hz
- 64 GB onboard eMMC boot storage
- Dual Gigabit Ethernet interfaces
- 12–20 V DC power input
- Compact 121 × 120 × 51 mm enclosure footprint
A key element is its Nash BPU AI accelerator, which provides up to 128 TOPS of AI computing performance. The platform also includes a six-core Cortex-A78AE processor and Cortex-R52+ processing resources for real-time tasks. Together, these elements allow the board to handle perception, decision-making and control workloads without necessarily requiring separate computing and control hardware.
The board is designed for applications involving machine vision, autonomous robotics, humanoid robots, robotic manipulators and intelligent edge systems. Its camera connectivity and expansion interfaces allow developers to connect imaging and peripheral hardware, making it suitable for systems that rely on visual perception and sensor inputs.
It also comes with 24GB of LPDDR5 memory and onboard storage, providing the capacity needed for larger AI models and robotics software. Connectivity options include Ethernet, USB, camera interfaces and expansion headers, giving developers flexibility when building prototype robotic systems and embedded AI equipment.
On the software side, the platform runs RDK OS and supports robotics development through ROS 2-related software and a collection of open-source algorithms and application examples. The ecosystem is intended to shorten the path from AI model development to deployment on physical robotic systems. With AI inference and real-time motion control integrated into one development platform, the RDK S100P targets the growing demand for compact computing hardware capable of supporting increasingly autonomous machines at the edge.
