Mobile Robot Taps Jetson Xavier Via New Aetina Carrier Board

By Eric Brown

Aetina announced a partnership to build an autonomous, solar-powered mobile robot with GPS tracking, sensors, and 6x HD cameras, based on its “AX710” carrier for the Linux-driven Jetson AGX Xavier.

Taiwan-based Aetina and an undisclosed third party are developing the UGV (Unmanned Ground Vehicle) robot for border and shore patrol and other remote inspection and exploration applications. The robot will be built around Nvidia’s powerful, AI-enabled Jetson AGX Xavier module via Aetina’s new AX710 carrier board.

AX710, front and back
(click images to enlarge)

We missed the AX710 when it was announced in February, but have detailed it farther below. The AX710, which follows Aetina’s earlier ACE-N310 carrier for the Jetson TX1/TX2/TX2i modules, joins CTI’s Rogue and Mimic Adapter carriers for the Xavier.

Nvidia’s 105 x 87 x 16mm Jetson AGX Xavier module has greater than 10x the energy efficiency and more than 20x the performance of the Jetson TX2, claims Nvidia. The module is equipped with 8x ARMv8.2 cores and a high-end, 512-core Nvidia Volta GPU with 64 tensor cores with 2x Nvidia Deep Learning Accelerator (DLA) engines. The Xavier is further equipped with a 7-way VLIW vision chip, as well as 16GB 256-bit LPDDR4 RAM and 32GB eMMC 5.1.

The upcoming UGV built around Aetina’s AX710 board will use the deep learning algorithms running on the Xavier to “understand and react to the surroundings” in real time based on camera and environmental sensor input, says Aetina. The company’s chief partner in the project will provide a GPS-based vehicle tracking system.

The robot will be equipped with solar panels. Their facing orientation will be continually adjusted to the best solar zenith angle calculated by the Xavier with the help of an optical sensor. Other sensors include thermal, infrared sensor, and metal detection sensors.

The UGV will be equipped with 6x Full HD, 360-degree cameras. Other features include LED lighting and the third-party, cloud-connected GPS tracking system, which features geofencing. The system will support up to 5G cellular connectivity as well as Innodisk’s iCAP remote monitoring system.

AX710 carrier

Like CTI’s Rogue carrier, Aetina’s compact, 112 x 107mm AX710 is designed to stack nicely with the 105 x 87 x 16mm Xavier module for deployment on robots and other space-constrained devices. Available with -25 to 80°C or -40 to 85℃ support with 10% to 90% humidity tolerance, the carrier board has a 9-20V DC input.

AX710 block diagram and Jetson AGX Xavier
(click images to enlarge)

The AX710 carrier provides coastline ports including 2x HDMI, 2x USB 3.1, and single USB 3.1 Type-C and micro-USB ports. There’s also a GbE port via RJ45, as well as two more GbE ports via an extension adapter.

AX710 portside view

The AX710 is equipped with 2x CANBus, 2x UART, 5x GPIO, and single RS-232, I2C, and “front panel” interfaces. There are also 4x I-PEX connectors, as well as M.2 M-Key and M.2 E-Key slots.

The board also provides a 60-pin extension slot. It’s unclear if that is the source of the MIPI-CSI-2 connections for Aetina’s optional camera modules. Although the robot will offer 6x HD cameras, the AX710 supports up to 8x HD cams. Alternatively, you can have 4x 4K cameras or an undisclosed number of GSML/FPD-LINK III SerDes(PDF) cameras.

The AX710 is available with an “iNAVI” optimization service featuring an embedded OS based on Linux, including secure boot and system recovery features. Aetina pre-integrates the necessary driver porting and and can customize OS specifications. Other technical services are also available.

Further information

No ship date or name was listed for the upcoming AX710-based UGV. Aetina Corporation will share a booth with its UGV partner at DSEI (Defence & Security Equipment International) at the Exhibition Centre London UK (EXCEL) from Sep. 10-13 at booth S4-205.

The AX710 carrier board appears to be available at an undisclosed price. More information may be found on Aetina’s AX710 product page.

This article originally appeared on on August 20.

Aetina |

4-Channel Automotive PMIC Meets Vehicle Camera Needs

Maxim Integrated Products has introduced a compact MAX20049 power management IC (PMIC) that integrates four power supplies into a tiny footprint. The device offers many options to support various output voltages, while also providing fault mitigation by flagging faults and shifts in output voltages.

Automotive camera modules tend to be size-constrained, so designers are constantly in search of a power management solution that can pack the necessary power and functionality into a small form factor. The 4-channel MAX20049 power management IC is almost 30 percent more compact than competitive solutions and offers the highest efficiency among other quad-power power management ICs in its class, says Maxim.

The chip offers many options to support modules that need various output voltages for different mixes of sensors and serializers, enabling designers to make changes in layout as needed or to fine-tune the IC to meet specific application requirements. The MAX20049 provides fault mitigation, a feature required by designers to help flag faults and shifts in output voltages to ensure that the cameras are working as needed.


  • Small Solution Size: has a PCB footprint that is almost 30 percent smaller than that of the closest competitor (38 mm2 compared to 53.3 mm2)
    • 4 outputs (dual bucks and dual LDOs) in a 3 mm x 3 mm QFN package
    • Protection unavailable in competitive products include over-voltage protection, under-voltage lockout, external power good (PGOOD) signal and cycle-by-cycle current limit
  • High Efficiency: system efficiency at full load is 74 percent (versus 69 percent for discrete automotive solutions)
  • Thermal performance plus high efficiency contributes to overall optimized performance
  • Flexibility: the dual buck converters and low-noise LDO support a wide voltage input range from 4V to 17 V, enabling power-over-coax (POC), typically from 8 V to 10 V. There is also an option of using one of the buck converters as an intermediate supply for generating typical sensor + serializer rails
    • Optimizes thermal performance and minimizes coax inrush current at startup
    • Flexible sequencing and fixed output voltages supporting various image sensors
  • Fault Mitigation: flags faults and shifts in output voltages to ensure cameras are working as intended
    • Once an over- or under-voltage signal is detected, the PGOOD pin will assert low
    • Cycle-by-cycle current limit implemented by the respective converter if either output is shorted
  • Low Noise: spread spectrum and 2.2 MHz switching frequency mitigates electromagnetic interference (EMI) to meet CISPR low-noise specifications

Maxim Integrated |



A Drone By Any Another Name

Input Voltage

–Jeff Child, Editor-in-Chief


For over a decade before I joined the Circuit Cellar team, I was Chief Editor of a magazine that covered embedded computing technologies used in military systems. At that publication, I naturally wrote and edited a lot of articles about UAVs (Unmanned Aerial Vehicles). Just as am now, I was very particular about terminologies and what they represent. And one word I quite adamantly wouldn’t allow used in that magazine was “drone.” Back then I didn’t like the term for a number of reasons. First, drone is a word that implies mindlessness or lack of intelligence. To me that didn’t feel right when covering military UAVs, because they typically embedded massive amounts of computing. Large military UAVs like the Global Hawk even had full backplanes of FPGA-based boards to do processing of imaging data and other functions. A second reason is that within the defense electronics industry, UAV was the term preferred over drone. Drone was what the unknowing, non-industry public called them—the word used for them in news stories. Most news stories using the word drone were—often justifiably—bad news.

So, for those reasons I banished any use of the word drone in that publication—at least I did before a change started happening in drone world. It’s important to understand that there are very few areas where the defense industry is ahead of the commercial industry. One exception, however, is UAVs—for many years the defense industry was way out ahead of the commercial world in UAV technology and development activity.

But around 2014 or 2015 a shift happened where biggest growth area for drone technology became dominated by commercial/civil unmanned platforms. Within that the largest chunk is the huge number of small hobbyist kinds of air vehicles. But as commercial uses blossomed for drones—ranging from film making to agriculture to construction and more—the drone market morphed toward a multi-billion-dollar market.

With that trend happening, I softened my stance, and I did start using the term drone when referring to consumer and commercial drones. And I knew that the defense electronics industry in this day and age has to keep tabs on the consumer technology market, because that’s where the rapid innovations happen. It’s too soon to tell what impact the rapid growth of the commercial/consumer drone industry will have on the defense side of drone technology. And since most (but not all) military drones are fixed-wing and commercial drones are mostly (but not all) rotary-wing, they may continue down separate paths. But it will be important for the defense industry to keep its eyes on where commercial drone technology is going.

Interestingly, this transition from defense to commercial also played out in the tradeshow realm. When I was at that military publication, the AUVSI Unmanned Systems show was a key event that I attended every year—this was even before they shifted to the new name Xponential and then to AUVSI Xponential. That show was dominated by companies marketing to the defense UAV market, along with all the defense primes (their customers). But in the 2014/2015 time frame, that show transitioned to where the number of consumer and commercial drone companies exhibiting began to be in the majority, and that’s been its direction ever since. While that wasn’t a positive trend for me when I was covering defense technologies, it’s very much welcome for me here on Circuit Cellar.

I have to be honest, writing about consumer and commercial drones is way more fun than covering military drone technology. As I’ve said before in this column, drone technology fascinates me partly because it represents one of the clearest examples of an application that wouldn’t exist without today’s level of chip integration driven by Moore’s law. That high level of integration has enabled 4k HD video capture, image stabilization, new levels of autonomy and even highly compact supercomputing to fly aboard today’s commercial and consumer drones. I’m looking forward to attending this year’s AUVSI Xponential event in Chicago, and next month I’ll be sure to share with you my thoughts about what I saw there. And as far as my objections to the word drone? Clearly, I’m over it.

This appears in the April 345 issue of Circuit Cellar magazine

Not a Circuit Cellar subscriber?  Don’t be left out! Sign up today:

Drones Tap a Variety of Video Solutions

Eyes in the Skies

In one way or another, much of today’s commercial drone development revolves around video. Technology options range from single-chip solutions to complex networked arrays.

By Jeff Child, Editor-in-Chief

Commercial drones represent one of the most dynamic, fast-growing segments of embedded systems design today. And while all aspects of commercial drone technology are advancing, video is front and center. Because video is the main mission of the majority of commercial drones, video technology has become a center of gravity in today’s drone design decisions. But video covers a wide set of topics including single-chip video processing, 4k HD video capture, image stabilization, complex board-level video processing, drone-mounted cameras, hybrid IR/video camera and mesh-networks for integrated multiple drone camera streams.

Technology suppliers serving all of those areas are under pressure to deliver products to integrate into video processing, camera and communications electronics inside today’s commercial drones. Drone designers have to pack in an ambitious amount of functionality onto their platforms while keeping size, weight and power (SWaP) as low as possible. Feeding these needs, vendors at the chip, board and system-level continue to evolve their existing drone video technologies while also creating new innovative solutions.

Video Processing SOC

Exemplifying the cutting edge in single-chip video processing for drones, Ambarella in March introduced its CV2 camera SoC (Photo 1). It combines advanced computer vision, image processing, 4Kp60 video encoding and stereovision in a single chip. Targeting drone and related applications, the company says it delivers up to 20 times the deep neural network performance of Ambarella’s first generation CV1 chip. Fabricated in advanced 10nm process technology, CV2 offers extremely low power consumption.

Photo 1
The CV2 camera SoC combines advanced computer vision, image processing, 4Kp60 video encoding and stereovision in a single chip.

The CV2’s CVflow architecture provides computer vision processing up to 4K or 8-Megapixel resolution, to enable object recognition and perception over long distances and with high accuracy. Its stereovision processing provides the ability to detect generic objects without training. Advanced image processing with HDR (High Dynamic Range) processing delivers outstanding imaging even in low light and from high contrast scenes. Its highly efficient 4Kp60 AVC and HEVC video encoding supports the addition of video recording to drone platforms.

At the heart of the CV2 is a Quad-core 1.2 GHz ARM Cortex A53 with NEON DSP extensions and FPU. CV2 includes a full suite of advanced security features to prevent hacking, including secure boot, TrustZone and I/O virtualization. A complete set of tools is provided to help embedded systems developers easily port their own neural networks onto the CV2 SoC. This includes compiler, debugger and support for industry standard training tools including Caffe and TensorFlow, with extensive guidelines for CNN (Convolutional Neural Network) performance optimizations.

Board-Level Solutions

Moving up to the board-level, Sightline Applications specializes in onboard video processing for advanced camera systems. Its processor boards are designed to be integrated at the camera level to provide low-latency video processing on a variety of platforms including commercial drones. Sightline offers two low SWaP board products. Both products are supported by SLA’s Video Processing Software: a suite of video functions that are key in a wide variety of ISR applications. The processing software has two pricing tiers, SLE and SLA. SLE provides processing only and SLA processes the video and provides telemetry feedback. . …

Read the full article in the May 334 issue of Circuit Cellar

Don’t miss out on upcoming issues of Circuit Cellar. Subscribe today!
Note: We’ve made the October 2017 issue of Circuit Cellar available as a free sample issue. In it, you’ll find a rich variety of the kinds of articles and information that exemplify a typical issue of the current magazine.

FREE Sample Issue – Oct. 2017


We’ve made the October 2017 issue of Circuit Cellar available as a sample issue. In it, you’ll find a rich variety of the kinds of articles and information that exemplify a typical issue of the current magazine.

Don’t miss out on upcoming issues of Circuit Cellar. Subscribe today!

Inside This Issue:
Emulating Legacy Interfaces
Do it with Microcontrollers
By Wolfgang Matthes

Building a Retro TV Remote
PIC MCU-Based Design
By Dev Gualtieri
Building a Robot Hand
With Servos and Electromyography
By Michael Haidar, Jason Hwang and Srikrishnaa VadivelLogger Device Tracks Amp Hours (Part 1)
Measuring Home Electricity
By William Wachsmann

Commercial Drone Design Solutions Take Flight

Chips, Boards and Platforms
By Jeff Child

Design for Manufacturing: Does It Have to be so Difficult?
An interview with Scott N. Miller and Thos Niles
By Wisse Hettinga

Signal Chain Tech Pushes Bandwidth Barriers
ADCs, FPGAs and DACs
By Jeff Child

Embedded in Thin Slices
Build an Embedded Systems Consulting Company (Part 6)
Trade-Offs of Fixed-Price Contracts
By Bob JapengaThe Consummate Engineer
In the Loop on Positive Feedback
New Value in an Old Concept
By George Novacek


The Darker Side
Antenna Performance Measurement Made Easy
Covering the Basics
By Robert Lacoste
From the Bench
Gas Monitoring and Sensing (Part 1)
Fun with Fragrant Analysis
By Jeff BachiochiTECH THE FUTURE
The Future of PCB Design

Racing to Keep Pace With PCB Complexities
By Duane Benson

Don’t miss out on upcoming issues of Circuit Cellar. Subscribe today!

Commercial Drone Design Solutions Take Flight

Chips, Boards and Platforms

The control, camera and communications electronics inside today’s commercial drones have to pack in an ambitious amount of functionality while keeping size, weight and power as low as possible.

By Jeff Child, Editor-in-Chief

There aren’t many areas of embedded systems these days that are as dynamic and fast-growing as commercial drones. Drones represent a vivid example of a technology that wouldn’t have been possible if not for the ever-increasing levels of chip integration driven by Moore’s law. Drones are riding that wave, enabling an amazing rate of change so that 4k HD video capture, image stabilization, new levels of autonomy and even highly integrated supercomputing is now possible on drones.


The Intel Aero Ready to Fly Drone is a pre-assembled quadcopter built for professional drone application developers. The platform features a board running an Intel 2.56 GHz quad-core Intel Atom x7-Z8750 processor.

The Intel Aero Ready to Fly Drone is a pre-assembled quadcopter built for professional drone application developers. The platform features a board running an Intel 2.56 GHz quad-core Intel Atom x7-Z8750 processor.

To get a sense of the rapid growth of drone use, just consider drones from the point of view of the Federal Aviation Administration (FAA). Integrating commercial drones into the FAA’s mission has been a huge effort over the past couple years. To paraphrase Michael P. Huerta, Administrator of the FAA, there are over 320,000 registered manned aircraft today and it took 100 years to reach that number. In contrast, only nine months after the FAA put its drone registration process in place, there were more than 550,000 registered users—comprised of both hobbyists and commercial drone users.

Electronics for Drones

Today’s commercial/civilian drone technologies are advancing faster than most people could have imagined only a couple years ago. And drone designs will continue to reap the benefits of advances in processor / chip technologies, sensor innovations and tools that make them easier to create. Feeding those needs, chip and board vendors of all sizes have been rolling out solutions to help drone system developers create new drone products and get to market quickly. Among these vendors are large players like Intel and Qualcomm–along with a whole host of specialized technology suppliers offering video ICs, single-chip cameras and a variety of sensor solutions all aimed at drone platforms. ….

We’ve made the October 2017 issue of Circuit Cellar available as a sample issue. In it, you’ll find a rich variety of the kinds of articles and information that exemplify a typical issue of the current magazine.
Don’t miss out on upcoming issues of Circuit Cellar. Subscribe today!


A Workspace for “Engineering Magic”


Photo 1—Brandsma describes his workspace as his “little corner where the engineering magic happens.”

Sjoerd Brandsma, an R&D manager at CycloMedia, enjoys designing with cameras, GPS receivers, and transceivers. His creates his projects in a small workspace in Kerkwijk, The Netherlands (see Photo 1). He also designs in his garage, where he uses a mill and a lathe for some small and medium metal work (see Photo 2).


Photo 2—Brandsma uses this Weiler lathe for metal work.

The Weiler lathe has served me and the previous owners for many years, but is still healthy and precise. The black and red mill does an acceptable job and is still on my list to be converted to a computer numerical control (CNC) machine.

Brandsma described some of his projects.


Photo 3—Some of Brandsma’s projects include an mbed-based camera project (left), a camera with an 8-bit parallel databus interface (center), and an MP3 player that uses a decoder chip that is connected to an mbed module (right).

I built a COMedia C328 UART camera with a 100° lens placed on a 360° servomotor (see Photo 3, left).  Both are connected to an mbed module. When the system starts, the camera takes a full-circle picture every 90°. The four images are stored on an SD card and can be stitched into a panoramic image. I built this project for the NXP mbed design challenge 2010 but never finished the project because the initial idea involved doing some stitching on the mbed module itself. This seemed to be a bit too complicated due to memory limitations.

I built this project built around a 16-MB framebuffer for the Aptina MT9D131 camera (see Photo 3, center). This camera has an 8-bit parallel databus interface that operates on 6 to 80 MHz. This is way too fast for most microcontrollers (e.g., Arduino, Atmel AVR, Microchip Technology PIC, etc.). With this framebuffer, it’s possible to capture still images and store/process the image data at a later point.

This project involves an MP3 player that uses a VLSI VS1053 decoder chip that is connected to an mbed module (see Photo 3, right). The great thing about the mbed platform is that there’s plenty of library code available. This is also the case for the VS1053. With that, it’s a piece of cake to build your own MP3 player. The green button is a Skip button. But beware! If you press that button it will play a song you don’t like and you cannot skip that song.

He continued by describing his test equipment.


Photo 4—Brandsma’s test equipment collection includes a Tektronix TDS220 oscilloscope (top), a Total Phase Beagle protocol analyzer (second from top), a Seeed Technology Open Workbench Logic Sniffer (second from bottom), and a Cypress Semiconductor CY7C68013A USB microcontroller (bottom).

Most of the time, I’ll use my good old Tektronix TDS220 oscilloscope. It still works fine for the basic stuff I’m doing (see Photo 4, top). The Total Phase Beagle I2C/SPI protocol analyzer Beagle/SPI is a great tool to monitor and analyze I2C/SPI traffic (see Photo 4, second from top).

The red PCB is a Seeed Technology 16-channel Open Workbench Logic Sniffer (see Photo 4, second from bottom). This is actually a really cool low-budget open-source USB logic analyzer that’s quite handy once in a while when I need to analyze some data bus issues.

The board on the bottom is a Cypress CY7C68013A USB microcontroller high-speed USB peripheral controller that can be used as an eight-channel logic analyzer or as any other high-speed data-capture device (see Photo 4, bottom). It’s still on my “to-do” list to connect it to the Aptina MT9D131 camera and do some video streaming.

Brandsma believes that “books tell a lot about a person.” Photo 5 shows some books he uses when designing and or programming his projects.


Photo 5—A few of Brandsma’s “go-to” books are shown.

The technical difficulty of the books differs a lot. Electronica echt niet moeilijk (Electronics Made Easy) is an entry-level book that helped me understand the basics of electronics. On the other hand, the books about operating systems and the C++ programming language are certainly of a different level.

An article about Brandsma’s Sun Chaser GPS Reference Station is scheduled to appear in Circuit Cellar’s June issue.