May 28, 2019      

NEW YORK – DJI today announced Manifold 2, an ultra-compact onboard supercomputer for the company’s aerial drones that provides additional computing capability in order for the drones to fly autonomously while identifying objects and avoiding obstacles. In addition, by adding Manifold 2 to a drone, users can process complex image data onboard the drone and get results immediately, said DJI, an RBR50 2019 winner.

DJI Manifold 2 supercomputer module

DJI Manifold 2. Image: DJI

The Manifold 2 is now available through authorized DJI Enterprise resellers worldwide. For North American customers, the system can be purchased through DJI’s online store, with the CPU version costing $1,379, and the GPU version at $1,099.

“Through partnerships with some of the industry’s leading computing companies and chip manufacturers, we are enabling DJI drones to become intelligent flying robots that can perform complex computing tasks and advanced image processing literally on the fly,” said Jan Gasparic, DJI director of strategic partnerships. “Manifold 2 expands the use cases of drone technology exponentially by allowing businesses, developers and researchers to build the most advanced autonomous aerial robotics solutions for nearly any industry or task.”

Users can choose between an Intel Core i7-8550U CPU or an NVIDIA Jetson TX2 supercomputer as the module for Manifold 2, enabling AI at the edge, DJI said. Both versions are compatible with the company’s enterprise drones, including the Matrice 210 series and Matrice 600 series (see video, above), and can be integrated into the drone’s internal systems and sensors using DJI’s software development kit. Each Manifold 2 includes USB, UART, CAN ports and more, giving developers the chance to link multiple Manifold 2 processors and build on top of them. The system also works with the N3 Flight Controller and A3 Flight Controller, DJI said.

“DJI Manifold 2 not only enables the advanced computer vision, perception and navigation featured in Automodality’s patented AMROS and Perceptive Navigation platform suite, it also enables next-generation artificial intelligence out on the edge, a hallmark of Automodality’s software and service offering for industrial inspections,” said Ed Koch, chief technology officer at Automodality. “By empowering the drone to perceive and analyze in real-time, Manifold 2 opens new areas and capabilities and allows Automodality-equipped DJI Matrice 210s to autonomously fly where no one else has flown before, in GPS and radio frequency-denied environments, where the most valuable and actionable data resides.”