The Lily Camera developed using ROS

| No Comments | No TrackBacks

From Henry Bradlow

"Throw your Lily in the air like you just don't care" is a popular comment on YouTube for the new Lily Camera. The fully autonomous flying camera went viral on May 12th when it was announced by Lily Robotics. It focuses on filming you, so you can focus on your activity.

Many people in the robotics and film communities have predicted that a product like the Lily Camera would enter the market, given the recent popularity of using drones for filming. While several companies have attempted to develop similar products, no company has achieved the capacity or flexibility of the Lily Camera, and no company has earned such enthusiastic attention.

What distinguishes the Lily Camera is its ease of use. The user throws the camera into the air and the Lily Camera automatically follows the user, capturing shots that are unmatched by any other device. The engineers at Lily Robotics integrate strategies from robotics, computer vision, and signal processing to ensure that Lily always knows the location of itself and the user it is filming. To achieve this situational awareness, the Robot Operating System, ROS, is heavily utilized.

In Lily Camera prototypes, ROS was used for passing messages between the tracking device and the Lily Camera. According to Rowland O'Flaherty, Lily Robotics' lead controls engineering, "Based on the sheer nature of how ROS is structured, it is seamless to pass messages between different devices. Sometimes you even forget that there are separate devices communicating with each other."

ROS is also leveraged for the testing and development of the Lily Camera. The engineers at Lily Robotics use RViz, ROS's 3D visualization tool, to simulate the movements of the camera for development, to visualize live test flights for real-time analysis, and to replay test flights for debugging and examination. Rowland added, "We can either run a simulated flight or a real flight with a flip of a switch thanks to ROS. Both the simulator and the real robot run the same code (ROS nodes), which rapidly increases the development cycle." 

No TrackBacks

TrackBack URL:

Leave a comment

Find this blog and more at

Monthly Archives

About this Entry

This page contains a single entry by Tully Foote published on May 26, 2015 2:06 PM.

ROS Jade Turtle Release was the previous entry in this blog.

New skeleton tracker node for ASUS Xtion Pro Live is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.