InOrbit announced the release of its newest robot operations (RobOps) tool Data Backfill. The tool stores data locally so that it can fill gaps in an autonomous mobile robot’s (AMR’s) operating history.
With its latest tool, InOrbit wanted to solve a problem that existed specifically in robot deployments. In a lab, a robot will always have a stable network connection and stay relatively still. This means engineers don’t need to worry about how much or when they can send data from the robot to a computer.
In real deployments, robots will run into unstable network connections while also producing up to gigabytes of data every minute from camera images, laser scans, maps, temperature sensors and wheel odometry, among other things. Lost connections can result in gaps of crucial data.
Data Backfill captures all of the data that a robot collects, even if the robot is offline, it then uploads that data to the cloud when the robot is able to connect again. Operators can even turn off robot data transmission and turn it on later, like when the robot is docking or charging.
InOrbit has already tried to help operators get the most out of their data with Adaptive Diagnostics, its program that instructs robots to transmit only the minimum information necessary for useful insights. Adaptive Diagnostics allows InOrbit to throttle data up or down, depending on the situation, but Data Backfill takes this idea further to provide more support.
Along with Data Backfill, InOrbit offers Time Capsule, a tool that lets AMR operators better understand where and why robot failures occurred. Released last year, the feature works well alongside Data Backfill to give operators more insight into their robot data even in challenging circumstances.
In February 2022, InOrbit released Configuration as Code, which added additional capabilities and best practices for RobOps. The release included version control of code changes, improved workflows, CI/CD integration and rollback capabilities, among other things.