Archive

Posts Tagged ‘sdk’

NVIDIA Xavier AI SoC Now Sampling, DRIVE IX & DRIVE AR SDKs Announced

January 8th, 2018 1 comment

Well over a year ago, NVIDIA introduced Xavier, their next generation self-driving and artificial intelligence processor, with eight custom ARM cores, a 512-core Volta GPU, and support for 8K video encoding and decode. A few months ago, the company provided some more details and unveiled NVIDIA DRIVE PX Pegasus A.I. computer for level 5 autonomous driving with two Xavier processors and two NVIDIA next-generation GPUs delivering a total 320 TOPS of computing power. For that it’s worth, 320 TOPS is about 3200 times more powerful than Intel Movidus Neural Network Compute Stick.

CES 2018 has now started, and NVIDIA made several announcement related to gaming and automotive markets, and confirmed Xavier is now sampling to select customers.

Click to Enlarge

What’s really new from the announcement is the addition of two new SDKs (software development kits) for the processor beside the original NVIDIA DRIVE AV autonomous vehicle platform:

  • DRIVE IX – Intelligent experience software development kit that will enable AI assistants for both drivers and passengers using sensors inside and outside the car.
  • DRIVE AR – Augmented Reality SDK designed for interfaces that deliver information points of interest along a drive, create alerts and navigate safely and easily.

This type of powerful hardware and software is however reserved to automotive customers, so most of us won’t be able to get hold of such platform, but we may end up being users of the technology soon enough, as NVIDIA announced partnerships with Volkswagen, Uber, ZF tier-one automotive supplier working with Baidu, and Aurora, a US startup designing and building self-driving technology.

Firefly ROC-RK3328-CC Development Board Now for Sale for $35 and Up

January 8th, 2018 8 comments

Remember Libre Computer Renegade SBC – aka ROC-RK3328-CC – based on Rockchip RK3328 processor, and launched last month on Indiegogo? The crowdfunding campaign successfully completed a few days ago with over $20,000 raised from close to 250 backers.

When we zoomed on the board we could see markings on the board indicated that it was made by Firefly team, the makers of boards such as Firefly-RK3288 Reload or Firefly-RK3399, and the company is now selling the board with 1 or 2GB RAM for respectively $35 and $50 plus shipping, but the 4GB version is not up for sale at this stage.

Firefly-ROC-RK3328-CC board specifications:

  • SoC – Rockchip RK3328 quad core Cortex A53 processor with ARM Mali-450MP2 GPU
  • System Memory – 1, 2, or 4GB DDR4
  • Storage – eMMC 5.x flash module socket (8 to 128 GB) + micro SD card slot
  • Video & Audio Output – HDMI 2.0a up to 4K @ 60 Hz with HDR10 and HLG support, 3.5mm AV port (composite video + stereo audio)
  • Video Codec – 4K VP9, H.265 and H.264, 1080p VC-1, MPEG-1/2/4, VP6/8
  • Connectivity – Gigabit Ethernet
  • USB – 2x USB 2.0 ports, 1x USB 3.0 port
  • Expansion Headers
    • 40-pin (mostly) Raspberry Pi compatible GPIO header with PWM, I2C, SPI, GPIOs
    • 3-pin ADC Header with 2x analog inputs, GND
  • Debugging – UART header pins
  • Misc – IR receiver; button
  • Power Supply – 5V via micro USB port
  • Dimensions –  85 x 56 mm

The board will ship with a micro USB cable, which will mitigate some of the micro USB powering issues some people may have had otherwise.

The company provides Ubuntu 16.04 32- & 64-bit, Debian 32-bit,  Android 7.1.2 firmware images, as well as the Android SDK and tools for download (Baidu links only for now). Android & Linux software and hardware documentation is also available.

Thank to Fran for the tip.

Xiaomi Unveils Their Very Own ESP32 Development Board, Module & SDK

November 28th, 2017 11 comments

We now have plenty of modules and board based on Espressif System ESP32 dual core wireless SoC, and a fairly good software support with ESP IDF SDK, Arduino CoreMicroPython, Espruino and other solutions.

So I was surprised when I saw Teo Swee Ann, Espressif CEO, posted photos about an event – Xiaomi IoT Developer Conference – where Xiaomi apparently introduced their own ESP32 development board, together their own SDK (still empty right now) compatible with the company’s Mijia ecosystem, as part of their “Open IoT platform”

Click to Enlarge

The slide below translates as “WIFI module –  Automated manufacturing / 4-fold foolproof mechanism / whole process tracing / cost price (about 15 RMB)” and “Open source SDK”, with the price likely the price of Xiaomi ESP32 module (ESP-WROOM-32C) shown on board above (Thank to hm for translation).

Click to Enlarge

I could not find much information at this early stage, except new partnership between Baidu and Xiaomi for A.I. and IoT which does not mention the board. I’ll update this post, once more details are available.

RedRat-X IR, Bluetooth & RF4CE Box Can Control Multiple TVs and Set-top Boxes

October 26th, 2017 1 comment

IR blasters can be used to control multiple IR devices through infrared, or other interfaces, for example DVR can use such device to change the channel of a set-top box just before recording.

RedRat-X is one of those devices for automating control of TVs, STBs and other infrared equipment, but it’s quite more versatile, as beside its built-in IR blaster, it adds 3 IR outputs were you can connect your own IR transmitters, as well as USB and Ethernet for remote controls. Furthermore, it can also take add-on modules to emulate  Bluetooth and RF4CE remote controls.

RedRat-X specifications:

  • IR blaster via the front of the unit. 0 mA to 250 mA in 100 steps.
  • 3x plug-in IR jack sockets which can be used in two modes:
    • Current mode (default): For use with plug-in IR flashers – 0mA to 100mA in 100 steps.
    • Voltage mode: To interface to IR distributions systems, such as Xantech, Buffalo etc.
  • Optional add-on modules – Bluetooth or RF4CE functionality
  • Control Interfaces – USB or Ethernet
  • Misc – Multicolor indicator LEDs
  • Power Supply – 5V via USB
  • Power Consumption – <= 2W in normal operation.
  • Dimensions – 11.5cm x 7.5cm x 2.5cm

The company (RedRat) also offers a range of software solutions to control and manage their IR blasters including RedRat Device Manager for initial setup, configuration and firmware updates, IR Signal Database Utility to record your existing remote control codes, and RedRat’s Scheduler application to automate your setup.

Click to Enlarge

In order to get even more control, TestManager application provides script based control of STBs, and an NET SDK working on Windows (using .NET), as well as Mac OS and Linux machines (using Mono). The company also mentioned RedRatHub application, which provides a socket based mechanism to send instructions to RedRat hardware with sample client code written in Python, C#, Perl, or PHP. The company only advertises the solution for audio-visual applications, but it might be possible to integrate it into home automation system to control air conditioners, door bells, light bulbs, etc…

RedRat-X is specialized hardware with the box selling for £250 ex VAT ($331 US). You’ll find more details about hardware, software, and documentation on the product page. You might be possible to reproduce something similar with a Raspberry Pi 3 + infrared HAT, and optional Bluetooth and RF4CE USB dongles, but then all the software part would be on you.

Via Electronics Weekly

NVIDIA DRIVE PX Pegasus Platform is Designed for Fully Autonomous Vehicles

October 11th, 2017 1 comment

Many companies are now involved in the quest to develop self-driving cars, and getting there step by step with 6 levels of autonomous driving defined based on info from  Wikipedia:

  • Level 0 – Automated system issues warnings but has no vehicle control.
  • Level 1 (”hands on”) – Driver and automated system shares control over the vehicle. Examples include Adaptive Cruise Control (ACC), Parking Assistance, and Lane Keeping Assistance (LKA) Type II.
  • Level 2 (”hands off”) – The automated system takes full control of the vehicle (accelerating, braking, and steering), but the driver is still expected to monitor the driving, and be prepared to immediately intervene at any time. You’ll actually have your hands on the steering wheel, just in case…
  • Level 3 (”eyes off”) – The driver can safely turn their attention away from the driving tasks, e.g. the driver can text or watch a movie. The system may ask the driver to take over in some situations specified by the manufacturer such as traffic jams. So no sleeping while driving 🙂 . The Audi A8 Luxury Sedan was the first commercial car to claim to be able to do level 3 self driving.
  • Level 4 (”mind off”) – Similar to level 3, but no driver attention is ever required. You could sleep while the car is driving, or even send the car somewhere without your being in the driver seat. There’s a limitation at this level, as self-driving mode is limited to certain areas, or special circumstances. Outside of these areas or circumstances, the vehicle must be able to safely park the car, if the driver does not retake control.
  • Level 5 (”steering wheel optional”) – Fully autonomous car with no human intervention required, no other limitations

So the goal is obviously to reach level 5, which would allow robotaxis, or safely drive you home whatever your alcohol or THC blood levels. This however requires lots of redundant (for safety) computing power, and current autonomous vehicle prototypes have a trunk full of computing equipments.

NVIDIA has condensed the A.I processing power required  or level 5 autonomous driving into DRIVE PX Pegasus AI computer that’s roughly the size of a license plate, and capable of handling inputs from high-resolution 360-degree surround cameras and lidars, localizing the vehicle within centimeter accuracy, tracking vehicles and people around the car, and planning a safe and comfortable path to the destination.

The computer comes with four A.I processors said to be delivering 320 TOPS (trillion operations per second) of computing power, ten times faster than NVIDIA DRIVE PX 2, or about the performance of a 100-server data center according to Jensen Huang, NVIDIA founder and CEO. Specifically, the board combines two NVIDIA Xavier SoCs and two “next generation” GPUs with hardware accelerated deep learning and computer vision algorithms. Pegasus is designed for ASIL D certification with automotive inputs/outputs, including CAN bus, Flexray, 16 dedicated high-speed sensor inputs for camera, radar, lidar and ultrasonics, plus multiple 10Gbit Ethernet

Machine learning works in two steps with training on the most powerful hardware you can find, and inferencing done on cheaper hardware, and for autonomous driving, data scientists train their deep neural networks NVIDIA DGX-1 AI supercomputer, for example being able to simulate driving 300,000 miles in five hours by harnessing 8 NVIDIA DGX systems. Once trained is completed, the models can be updated over the air to NVIDIA DRIVE PX platforms where inferencing takes place. The process can be repeated regularly so that the system is always up to date.

NVIDIA DRIVE PX Pegasus will be available to NVIDIA automotive partners in H2 2018, together with NVIDIA DRIVE IX (intelligent experience) SDK, meaning level 5 autonomous driving cars, taxis and trucks based on the solution could become available in a few years.

Hologram Unveils Nova 3G USB Dongle and Python SDK; 200 Raspberry Pi Zero W Kits Given Away to Developers

October 6th, 2017 No comments

This summer I discovered Hologram global cellular IoT SIM card, and since they provided free developer samples with 2MB of monthly data includes, I decided to get one to try it out. I received it a few weeks later, and to my surprise it worked, despite my country of residence having some strict requirements with regards to SIM card registration. The SIM card uses roaming, but with a low fixed worldwide pricing, and does not come with a phone number by default, so maybe that’s why I did not have to register.

The company is now back with Nova, an open source hardware cellular modem certified by OSHWA (ID #US000077). It’s basically 2G/3G USB dongle that’s controlled by Hologram Python SDK, specifically suited to Debian systems like Raspberry Pi 3 or BeagleBone Black. Hackster.io is also involved in the launch with a worldwide contest offering 200 free kits comprised of Nova 3G USB dongle and Raspberry Pi Zero W board for the best project ideas leveraging cellular IoT.

Nova will eventually come in three versions

  • 3G (in production now) – Ublox Sara-U201 module;  Global 3G/2G GSM;  GPRS/GSM/UMTS/HSPA: 850, 900, 1800, 1900 MHz;
  • Cat-M1 (November 2017) – Ublox Sara-R404M module; USA LTE Cat M-1; FDD: 13 (Verizon)
  • Cat-M1/NB1 (Q1 2018) – Ublox Sara-R410M-02B module; Global LTE Cat M-1+ NB; FDD: 1,2,3,4,5,8,12,13,17,18,19,20,25,26,28

All should have the same other interfaces and other specs:

 

  • u.FL Antenna Connector
  • Nano SIM card holder
  • UART GPIO Pads
  • USB Serial
  • Network Status LED; Power LED
  • Fully end certified (FCC, PTCRB, CE, and AT&T)
  • Dimensions – 46mm x 19mm x 6mm (Plugged in PCB);  71mm x 23mm x 9mm (w/ case)
  • Weight – 8 grams

The hardware kit includes the dongle, Hologram global IoT SIM card, a transparent enclosure, 2 Quad-band flexible u.FL antennas, and access to Hologram Developer Tools for modem and data management.

 

The dongle can be controlled using Hologram client tool, or Hologram Python SDK requiring ppp and Python 2.7 packages, and will allow you to send SMS, setup data connection, and more. Any SIM card should work, and it’s not tied to Hologram SIM card. While the company claims OSHWA certifications, the number US000077 is not present (empty line) in the OSHWA certification list yet, and so far, they’ve only released the PDF schematics. However, Python SDK is fully open source and released under an MIT license on Github.

More details can be found in the product page, and Nova 3G kit can be purchased now for $49.

But as mentioned in the introduction, if you have a great project idea, you could also get the kit for free, and possibly another “grand prize” (Apple Watch Series 3)once the project is completed. The contest is opened worldwide (except to US sanctioned countries) with the following timeline:

  • Submit your proposal by October 27, 2017
  • Best project ideas will be selected, and be sent their kit within around 14 days
  • Build and submit your project to Hackster.io by January 5, 2018
  • 8 Grand Prize winners will be announced on January 8, 2018 for four categories: gateway, asset tracking, remote controlling, and remote monitoring.

There are already 135 participants. Good luck!

 

STMicro BlueNRG-MESH SDK for Bluetooth Mesh to Include Code for Firmware, Android and iOS Apps

October 2nd, 2017 2 comments

Earlier this summer, the Bluetooth SIG announced Bluetooth Mesh, which supports many-to-many (m:m) device communications for up 32,767 unicast addresses per mesh network (in theory), and is compatible with Bluetooth 4.0 or greater hardware.  Several companies immediately unveiled Bluetooth Mesh SDK at the time including Qualcomm, Nordic Semi, and Silicon Labs.

ST Micro has now unveiled their own BlueNRG-MESH SDK which the company claims is “the market’s only three-part SDK that provides two app developer packages for Android and iOS, and the embedded-development software for building smart objects such as light fittings and sensors”.

Sadly, details about the SDK are near inexistent now, except – as one would expect – BlueNRG-MESH SDK will work with ST BlueNRG Bluetooth low energy wireless network processor based on an ARM Cortex M0 core, and corresponding development kits. [Update: STSW-BNRG-Mesh page has many more details about the SDK including the architecture diagram below.

]

The solution was showcased at Bluetooth Asia in Shenzhen last week, with some selected customers already working with the SDK, before the public release scheduled for Q4 2017.

Thanks to Jon for the tip.

TECHBASE Moduino X Series Industrial IoT Modules / Endpoints are Based on ESP32 WiSoC

September 27th, 2017 4 comments

We’ve previously covered TECHBASE ModBerry industrial IoT gateways leveraging Raspberry Pi 3, FriendlyELEC NanoPi M1 Plus, or AAEON’s UP Linux boards. The company has now launched Moduino X series modules powered by Espressif ESP32 WiFi + Bluetooth SoC to be used as end points together with their ModBerry gateways.

Moduino X1

Two models have been developed so far, namely Moduino X1 and X2, with the following specifications:

  • Wireless Module – ESP32-WROVER with ESP32 dual-core Tensilica LX6 processor @ 240 MHz, 4MB pSRAM (512KB as option), 4MB SPI flash;
  • External Storage – X2 only: micro SD card slot
  • Connectivity
    • 802.11 b/g/n WiFi up to 16 Mbps + Bluetooth 4.2 LE with u.FL antenna connector
    • X2 only: 10/100M Ethernet
    • Options: LoRa (Semtech SX1272); Sigfox (TI CC1125); LTE Cat M1/NB1; Zigbee
  • Serial – 2x RS-232/485
  • Display – Optional 0.96″ OLED display with 128×64 resolution
  • Expansion I/Os
    • 4x Digital I/O (0 ~ 3V)
    • 2x Analog Input:
    • A2 Only: 2x analog output (optional)
    • A2 only: support for Techbase ExCard add-on modules for extr RS-232/485 ports, Ethernet ports, PCIe slots, analog input and output, digital I/Os, relays, M-Bus interface, etc…
  • Battery – Optional battery power support (A1 only); optional UPS function with LiPo battery or Supercapacitor
  • Power Supply -5V DC
  • Dimensions
    • A1 – ABS: 90 x 36 x 32 mm (LxWxH); Aluminum: 95 x 35 x 41 mm (LxWxH)
    • A2 – ABS: 90 x 71 x 32 mm (LxWxH); Aluminum: 95 x 71 x 41 mm (LxWxH)

Moduino A1 consumes less than A2, and can be powered by batteries only, but both models can use battery as UPS. The modules support Espressif ESP-IDF SDK, Zephyr Project, Arduino programming, MicroPython, Mongoose OS, and more, and would typically be used as meters & sensor nodes capable of reporting temperature, humidity, pressure, acceleration, & light with attached sensors. More sensors are being developed by the company.

Moduino X2 (right)

Moduino X1 & X2 appear to be available now, but you’d need to contact the company to get price information. Visit Moduino X series product page for more details.