Upbeat Unveils Bluemag Pi: An AI-Ready RISC-V Flight Controller

Overview

Upbeat Technology has officially launched the Bluemag Pi, a sensor-integrated flight controller that pairs real-time flight-critical performance with onboard AI inference. Built for long-endurance fixed-wing aircraft, commercial multirotor drones, and unmanned surface vessels, the board represents a significant shift in how autonomous systems handle complex computational loads.

The device runs on Upbeat’s UP301 system-on-chip, which features a dual-core architecture combining SiFive’s E3 and E2 RISC-V processor cores. This design allows developers to separate time-critical flight tasks from communication, navigation, and system-management functions, thereby improving overall responsiveness and reliability.

By integrating these capabilities into a single board, Upbeat aims to close the gap left by legacy autopilots, which typically require developers to bolt on a separate companion computer to achieve onboard intelligence. The Bluemag Pi is now available for purchase on the Upbeat Technology website.

Engineering and Architecture

At the heart of the Bluemag Pi is the UP301 SoC, engineered specifically for flight-critical applications where hard real-time guarantees are non-negotiable. The dual-core RISC-V architecture dedicates one core to flight control while the other handles AI inference and system tasks, ensuring neither function compromises the other.

The board provides ample resources to support advanced flight operations, including generous RAM, nonvolatile storage, hardware floating-point capability, timers, and direct memory access (DMA). It runs the Apache NuttX real-time operating system and maintains compatibility with leading open-source flight-control stacks such as PX4, ArduPilot, and Betaflight.

"Flight control is one of the toughest tests for any processor architecture... Bluemag Pi proves RISC-V is ready for more than edge AI, it's ready for the real-time, safety-critical systems that autonomous flight depends on." — Martyn Stroeve, Vice President, SiFive

From a sensor perspective, the board incorporates two independent inertial measurement units (IMUs), a barometer, and dual global navigation satellite system (GNSS) receivers. Sensor-fusion software continuously cross-checks readings from these components to produce dependable estimates of altitude, attitude, and velocity, even if a single sensor drifts or fails.

Connectivity is handled via a standard MAVLink radio link for telemetry streaming, while the integrated four-in-one electronic speed controller ensures precise throttle response. Additionally, onboard storage continuously logs flight data to aid in diagnostics, tuning, and post-flight analysis.

Market Context and Industry Growth

The launch of the Bluemag Pi coincides with a period of rapid expansion in the global drone market. According to Grand View Research, the market was valued at approximately $96.4 billion in 2026 and is projected to reach $182.4 billion by 2033, growing at a compound annual rate of roughly 9.5%.

This growth is largely driven by the adoption of AI-enabled, long-endurance platforms in sectors such as agriculture, infrastructure inspection, logistics, and defense. However, the flight-controller market has struggled to keep pace, with many incumbent autopilots running on general-purpose processors lacking dedicated AI acceleration.

Bluemag Pi addresses this bottleneck by offering a unified solution that eliminates the need for additional hardware. This integration reduces both complexity and power consumption, which are critical factors for long-endurance missions.

Steven Yen, Founder and CEO of Drone Dynamics, highlighted the practical benefits of this approach:

"We chose Upbeat's Bluemag Pi flight control board over standard solutions because it's remarkably easy to integrate yet built with the real-time performance and robust feature set advanced drone development demands."

Availability and Demonstrations

Engineers looking to implement custom flight algorithms can utilize the standard Bluemag Pi board, which fully supports secondary development. For those seeking to design proprietary carrier boards, Upbeat offers the UP301 chip directly. The platform supports a wide range of autonomous modes, including manual, stabilized, altitude-hold, position-hold, mission-based, and return-to-home operations.

Upbeat will showcase the Bluemag Pi at CEATEC 2026 in Chiba, Japan, from October 13–16. Attendees can experience the board’s real-time performance and AI inference capabilities at booth 7H215-01. The company will also display its humanoid robotic hands and other new products at the event.

Furthermore, Upbeat is scheduled to attend TAITRONICS & AIoT Taiwan from October 20–22 at booth J0102. These exhibitions will provide industry professionals with hands-on opportunities to evaluate the hardware in live flight scenarios.

In addition to the Bluemag Pi, Upbeat’s recent work in robotics has garnered recognition. The company received the Robot Intelligent Mobility System Excellence Award for its MEMS Edge AI Module, which enables equipment to detect abnormal sounds and flag failing components before breakdowns occur. This sensing platform is now being extended to humanoid robotics to detect grip slippage and surface friction in real time.