Now all VOXL products have SDK 1.1.2 parity.

Jan 11, 2024 Posted by: Lauren Young

SDK 1.1.2 is now available on all VOXL products.

Happy New Year! We’ve been hard at work making big changes under the hood. Our latest VOXL SDK 1.1.2 is now available on all VOXL products - including our original Blue UAS Framework 1.0 VOXL line. Experience streamlined configuration and faster video processing like never before.

SDK 1.1.2 feature highlights:

APQ8096 and Qualcomm Flight RB5 5G Platform support

Now all VOXL and RB5 products run the same SDK across the board

v1.14 on Flight Core v1 and v2
Our Flight Core V1 and V2 images are now built from the same codebase as the PX4 version running on VOXL 2 SDSP ensuring predictable PX4 behavior across the product line.

Hardware-accelerated video compression in voxl-camera-server

For hires color cameras, voxl-camera-server can now publish either H265 or H264 compressed video over libmodal-pipe. Furthermore, it can compress two video streams at different resolutions simultaneously. This enables, for example, one SD video stream for RTSP alongside one HD video stream for recording to disk with the new voxl-record-video tool. Both at 30hz, accelerated in hardware! This enabled the next feature:

Always-on RTSP video and MAVLink camera protocol support

Out of the box, in hardware configurations with a color hires camera, voxl-streamer and voxl-mavcam-manager are enabled by default so that as soon as you connect to QGroundControl you get a live video feed with no additional configuration steps. Furthermore, the video snapshot and video record functions in QGroundControl will automatically appear and allow you to take full resolution JPEG snapshots and HD video recordings that get saved to disk onboard VOXL in flight.

Streamlined configuration

A primary focus has been to streamline MPA configuration process by including it in the SDK install script. A new package, voxl-configurator consolidates tools for managing the SKU, calibration files, and configuring MPA services. All of these tools are called during the SDK flashing script.

Other

Improved thresholding in voxl-calibrate-camera that’s less sensitive to lighting conditions
16-bit raw lepton thermal camera data pipe
Send VIO packets with quality=-1 to PX4 on VIO failure to utilize PX4’s new quality metric functionality and make for more graceful fallback to altitude flight mode.
Robustness improvements to libmodal_pipe

Run Five Simultaneous Neural Networks on VOXL 2 with TensorFlow Lite

Jul 06, 2022 Posted by: Lauren Young voxl-2

VOXL 2 can run five concurrent neural networks with TensorFlow Lite. TensorFlow Lite is an embedded, open source program that allows developers to run pre-trained models for machine learning or computer vision applications.

VOXL^® 2 is the Next Generation Blue UAS Framework 2.0 Autopilot

Apr 25, 2022 Posted by: Chad Sweet

VOXL<sup>®</sup> 2 is the Next Generation Blue UAS Framework 2.0 Autopilot

As a Blue UAS Framework manufacturer, ModalAI^® is committed to advancing the U.S. drone industry. In 2019, we partnered with the Defense Innovation Unit (DIU) to produce VOXL Flight, the founding autopilot of the Blue UAS Framework program; a program that curates interoperable, NDAA-compliant UAS components and software that provides options for Government and industry partners. VOXL Flight enabled over 300+ partners, and was recognized by the RBR50 and BIG AI awards for being an open, enabling technology for drones. Now that the U.S. drone industry has grown to over 60 U.S. based drone manufacturers and is integrating utilitarian use cases in everyday life, we are pleased to introduce the next generation Blue UAS Framework autopilot, VOXL 2.

VOXL 2 is the world’s smallest and most advanced autopilot built in the USA. At only 16 grams, VOXL 2 boasts more AI computing than any other autopilot on the market and offers four times the computing of the previous generation. VOXL 2 integrates a PX4 real-time flight controller with an 8-core CPU, a GPU and NPU that provide a combined 15 Tera Operations Per Second (TOPs), seven image sensors, and TDK IMUs and barometer. With support for Wi-Fi, 4G and 5G connectivity, VOXL 2 enables mission critical use cases with reliable connectivity and beyond-visual-line-of-sight (BVLOS) navigation. This smaller-than-a-business-card and lighter-than-a-double-a-battery supercomputer autopilot will enable the next generation of smaller, smarter, and safer drones.

Smaller Drones

VOXL 2 is the smallest, most advanced autopilot to date. At only 16 grams, VOXL 2 weighs less than the average double A battery. As part of the Blue sUAS 2.0 program, VOXL 2 condenses most required electronics into a single, credit-card sized PCB, making it SWAP-optimized to power the smallest drones.

Before VOXL 2, developers might take a “Frankenstein” approach to building a robot with individual PCB components which could negatively affect the size and weight of the robot. VOXL 2’s advanced onboard computing and SWAP-optimized design eliminates the time and complexity that comes with sourcing and synchronizing individual autonomy components together.

Smarter Autonomy

VOXL 2 advances the latest Qualcomm Flight RB5 5G Platform. With the latest Qualcomm QRB5165 processor and an integrated PX4 flight controller on the DSP, VOXL 2 achieves powerful heterogeneous computing capabilities. The QRB5165 processor boasts an 8-core CPU up to 3.091GHz, 8GB LPDDR5, and 15 TOPs of AI performance.

Powerful onboard computing paired with advanced imaging capabilities makes VOXL 2 the smartest Blue sUAS autopilot to date. VOXL 2’s seven concurrent image sensors, including MIPI, time of flight (TOF), stereo pair, and 8K30, enable advanced indoor and outdoor navigation including vision-based SLAM for AI and movement, GPS-denied navigation, obstacle detection and avoidance.

Finally, VOXL SDK runs on VOXL 2. Developers can quickly bring their projects up to speed with VOXL 2’s open development platform. VOXL SDK offers support for popular open source applications, such as PX4, Linux Ubuntu 18.04, ROS 1, ROS 2, Open CV, Docker, and TensorFlow Lite.

Safer Navigation

The core of safe autonomous navigation is an advanced communication and perception system that works symbiotically. VOXL 2’s seven concurrent image sensors combined with advanced computing from the QRB5165 deliver advanced computer vision mapping and navigation, ensuring accurate autonomous maneuvers based on real-time data.

VOXL 2’s wide connectivity options also open a plethora of safer navigation-based use cases. With a 5G add-on board, VOXL 2 achieves low latency, high bandwidth data transfer that enhances mission critical navigation, including beyond visual line-of-sight (BVLOS) to support safer, more reliable flight.

VOXL 2 Available Now

VOXL 2 is available for purchase now at https://modalai.com/voxl-2

Blog 