Advantech Launched AI Companion Computer for Drones

The new ASR-D501 module enables edge-based autonomous perception for unmanned aerial vehicles.

Updated on Sept. 30, 2026 in Robotics

A professional industrial drone hovers in the sky, highlighting a compact, matte-black metal processor module mounted on its chassis.
Advantech has launched the ASR-D501, a new hardware module designed to enable edge-based AI perception for autonomous unmanned aerial vehicles. AI Illustration. Upload story photo >

Live Poll

Does the emergence of more capable AI drone technology make you optimistic about the future?

Advantech has announced the ASR-D501, a compact AI companion computer designed to handle perception workloads locally on autonomous drones. The device is a shippable hardware module optimized for low-power flight applications.

Why it matters

By moving perception processing from the cloud to the device itself, the ASR-D501 allows autonomous drones to operate reliably in environments with limited connectivity. This development accelerates the integration of high-level AI tasks into the restricted power and weight budgets of aerial robotics.

The ASR-D501 delivers 12 TOPS of AI performance while operating within a sub-10W power envelope. The board weighs 50g without a heatsink and measures 100 mm by 60 mm, integrating five 4-lane MIPI-CSI interfaces for high-speed camera data.

The players

Advantech

A manufacturer of industrial computing hardware that provides embedded boards, automation systems, and edge AI solutions for global infrastructure.

Qualcomm

A developer of wireless telecommunications products and specialized system-on-chip architectures for mobile, automotive, and edge AI applications.

The details

The ASR-D501 utilizes the Qualcomm QCS6490 processor to handle perception workloads locally, reducing latency and cloud reliance. It provides native support for ROS 2—the Robot Operating System, a set of software tools for robot application development—enabling direct interoperability with the PX4 and ArduPilot flight-controller ecosystems. Additionally, the unit is engineered for harsh environments with an operating temperature range of -20°C to 70°C and a vibration resistance rating of 3.5 Grms.

Timeline

  1. October 1, 2026: Advantech announced the ASR-D501 computer.

  2. 2027: Advantech plans to launch mid-range Qualcomm CQ8-based solutions.

The Tech Race

The ASR-D501 seeks to standardize AI processing for the growing autonomous unmanned aerial vehicle sector. By aligning with the ROS 2 software ecosystem, Advantech is positioning its hardware to become a primary development target for drone builders competing against proprietary, closed-source flight platforms.

Drone developers can begin integrating the module into their flight platforms to enable advanced perception capabilities without needing external cloud connectivity. The system supports Ubuntu 24.04 LTS, providing a familiar software environment for engineers building new aerial robotic workflows.

The takeaway

Advantech is betting that localized AI processing will become the standard requirement for scalable autonomous flight. Developers should monitor the company's 2027 product roadmap to see how their mid-range Qualcomm CQ8-based solutions compare to the performance benchmarks of the current ASR-D501 release.

What happens next

Advantech has scheduled the expansion of its UAV computing portfolio to include mid-range Qualcomm CQ8-based solutions in 2027.

Further reading

For more on the latest advancements in autonomous systems, see our coverage on Robotics.

Live Poll

Does the emergence of more capable AI drone technology make you optimistic about the future?