Path: EDN Asia >> News Centre >> Industrial/Mil/Aero >> Flying blind: Can we now add autonomy to drones?
Industrial/Mil/Aero Share print

Flying blind: Can we now add autonomy to drones?

21 Mar 2016  | Junko Yoshida

Share this page with your friends

Movidius hopes its partnership with DJI will earn the start-up a genuine head-start in the drone market.

The start-up initially found itself in the spotlight when Google picked a Movidius chip in Project Tango. The project's mission was to allow app developers to create a user experience for indoor navigation, 3D mapping, physical space measurement, augmented reality and recognition of known environments. Earlier this year, Movidius also entered into a new agreement with Google to bring super-intelligent models—extracted from deep learning at Google's data centres—over to mobile and wearable devices.

Movidius CEO El-Ouazzane is counting on the DJI partnership to give Movidius the lead in the industry-wide race to enable drones to see.

El-Ouazzane said, "DJI has an amazing computer vision team. When it comes to bringing autonomy to drones, they are at least one year ahead of all other drone companies." He said, "DJI took a gamble on autonomy. In the drone space, DJI is what Tesla is to the automotive market."

Movidius business model

Asked about specifics in the collaboration, El-Ouazzane said, "We offer DJI the underlying vision processing unit and its library (for depth tracking, optical flow and others) that would allow DJI to develop their best possible algorithms." Computer vision algorithms DJI has developed for its own drones are DJI's own property, he added.


Figure 3: Movidius Development & Prototyping board. (Source: Movidius)

Working with Google and DJI, Movidius has supplied its SDK and vision processing unit only. The idea was to allow these two customers to develop their own computer vision algorithms.

But Movidius is also seeking customers who need computer vision algorithms for their own applications, El-Ouazzane said. In such cases, "We will license them our own computer vision algorithms."

Gunning for drone design wins

Drones are rapidly becoming one of the most hotly pursued new product segments among chip giants like Intel and Qualcomm. Intel, armed with its RealSense technology, recently acquired a German drone company Ascending Technologies. Qualcomm has been pushing Snapdragon Flight into drones.

Last year, DJI revealed research that applied the guidance system for "unique applications, including an aerial solution created at Fudan University in Shanghai that uses Intel processors to detect illegally parked cars from the air."

Calling it a guidance system based on Intel's X86, Movidius CEO said, "A ultra-low-power computer vision system is what separates men from boys, so to speak." He stressed, "We have embedded computer vision SoCs built from the ground up, rather than something based on the existing chips."

The Movidius Myriad 2 VPU is currently in production using a 0.28nm HPC process technology at Taiwan Semiconductor Manufacturing Co. "We can't take a gamble on a newer geometry," said the Movidius CEO, because so much is riding on the start-up's vision processing unit. Besides hardware design, the two other pillars of the company are a software development kit that includes tool chains, and the development of machine intelligence. The latter two occupy 75 per cent of Movidius' team—mostly R&D engineers.


Figure 4: Myriad 2 has a heterogeneous architecture designed from the ground up for vision processing. (Source: Movidius)

 First Page Previous Page 1 • 2 • 3 Next Page Last Page

Want to more of this to be delivered to you for FREE?

Subscribe to EDN Asia alerts and receive the latest design ideas and product news in your inbox.

Got to make sure you're not a robot. Please enter the code displayed on the right.

Time to activate your subscription - it's easy!

We have sent an activate request to your registerd e-email. Simply click on the link to activate your subscription.

We're doing this to protect your privacy and ensure you successfully receive your e-mail alerts.

Add New Comment
Visitor (To avoid code verification, simply login or register with us. It is fast and free!)
*Verify code:
Tech Impact

Regional Roundup
Control this smart glass with the blink of an eye
K-Glass 2 detects users' eye movements to point the cursor to recognise computer icons or objects in the Internet, and uses winks for commands. The researchers call this interface the "i-Mouse."

GlobalFoundries extends grants to Singapore students
ARM, Tencent Games team up to improve mobile gaming

News | Products | Design Features | Regional Roundup | Tech Impact