INTRODUCING IMPROVED WORKFLOWS AND MOBILITY FOR CONSISTENT AND COMPREHENSIVE DATA COLLECTION
BLOG POST | November 11, 2020
Agile mobile robots like Spot can collect limitless quantities of site data, turbocharging existing analysis tools and enabling teams to focus on action rather than observation. In Spot’s first year on the market, we’ve seen diverse teams in an array of industries put the robot to use. During this time, we worked closely with hundreds of Spot users to understand their application development workflow: how they attach sensors, analyze data, and integrate the robot into their existing systems. We identified common obstacles and mapped out an easier path to implementation.
Spot Release 2.1 acts on those insights and makes it easy for you to attach your own sensors, collect and save the data you care about, and integrate that data into your existing systems. With 2.1, we’re launching several features which make Spot immediately useful out of the box for autonomous data collection missions. These can be used by operations teams to repeatedly collect vital data in dangerous or remote sites.
New in this release:
- Easily automate image collection workflows and aim the PTZ camera in Autowalk to get detailed inspection photos, autonomously.
- Put your data in-context with user-defined origins and custom metadata.
- Download mission data to the tablet for easy integration into third party tools.
EASY SENSOR INTEGRATION
Attaching new image sensors, like off-the-shelf spherical or thermal cameras, is now as easy as editing an example script and installing its docker container onto the Spot CORE compute payload. The new image sources show up on Spot’s tablet controller, and users can trigger captures in both teloperation and the easy-to-use Autowalk autonomy system. Spot can now be used to collect training images for computer vision models, to visualize data and model output live on the tablet controller, and to capture data from custom non-visual sensors like gas detectors or laser scanners.
Spot now has the ability to attach metadata to images, associating them with the robot’s location, user-defined labels, or custom values such as GPS coordinates from an attached payload. This enables users to put their data in context, for example: combining site photos from multiple missions into a single view, sorting images by asset-ID, or collecting datasets for computer vision model training. We’ve doubled-down on standard data types (JPEG images, JSON and CSV metadata files) to eliminate integration bottlenecks and built a high-performance system for developers to write their own data streams into the robot’s logs.
Powerful tools aren’t powerful if operators can’t use them in the field, so we’ve streamlined the data collection workflow significantly. Users can capture data manually and autonomously in Autowalk and download it to the tablet’s SD card for easy off-robot use. Common actions and callbacks can be configured on the tablet for quick use during operation. We’ve also made numerous under-the-hood improvements to Spot’s industry-leading locomotion and autonomy, further enabling operators to focus on the job and not the robot.
SUMMARY OF NEW FEATURES
Improved data collection during teleoperation and Autowalk missions
- Capture images from robot cameras, Spot CAM, and third-party image sources, tag them with metadata, and download them to the tablet for easy viewing and sharing.
- When recording Autowalk missions on the tablet, point Spot at items of interest and view image data in real-time.
- Annotate image data with robot position or custom metadata from easy-to-write data acquisition (DAQ) plugins.
- Export data in standard types for easy importation into off-the-shelf analysis tools.
Enhanced robot behaviors
- Improved obstacle avoidance, slip handling and recovery, stairs performance, self-righting on sloped terrain and near obstacles, crawl stability, and stability standing up with payloads and on sloped terrain.
- New roll-over behavior for easier access to the battery compartment.
- Improved fiducial position measurements and multi-fiducial support in GraphNav maps.
Spot CORE and CORE AI pre-configured software improvement
- Remote desktop support via TigerVNC.
- Cockpit provides a graphical system administration interface.
- Docker and Portainer enable deployment and management of Docker containers as components using a simple web UI.
- Pre-configured network settings provide better support for WiFi dongles.
On-robot log access for developers
- To assist with debugging, application developers using Spot can now access some logs stored on the robot, including:
- Logs of API requests and response messages.
- Text log messages and binary data.
- Developers can now contribute their own data to the on-robot log.
Easier system administration
- Spot now collects logs from batteries, clears them, and includes battery log data in the system log bundle.
- Admin Console now provides a UI for rebalancing an imbalanced battery.
- Spot can now associate with 802.1x enterprise WiFi networks, including those which require certificates.
These new features in Release 2.1 unlock Spot’s full data collection potential and set the stage for exciting new capabilities coming early next year: self-charging and remote operation. Upgrade to 2.1 today or contact sales to put Spot’s game-changing technology to work quickly and reliably, right out of the box.