Introducing Leo Examples

Discover what new features for the stock Leo Rover await you!

May 10, 2023


Aleksandra Szczepaniak

Leo Rover's object detection

We’ve developed a Leo Examples package containing three new features for the stock Leo Rover. The best part is that they’re free and available to everybody, and you don’t need any prior knowledge to run them on your robot. Let’s see what the features are :).

Line follower

Leo Rover's line follower example
A Leo Rover autonomously following a designated route

The major feature among the 3 ones within the Leo Examples package is line follower.

A line follower feature allows a robot to autonomously follow a visual line placed on a surface through a feedback mechanism that governs its movements. 

The operation of our line follower example lies in a neural network which we had to train. To enhance the model’s speed on the robot,we converted it using TensorFlow Lite. What distinguishes our line follower example is that it relies on not one, but two lines that designate a path for your Leo Rover to follow by itself. For the feature to work, you’ll need a Leo Rover, an insulating tape tomark a two-lined track, and the software uploaded to your robot.

To run the line follower, go through our tutorial. Once you’ve done that, your Leo Rover will autonomously navigate the route you designed.

Learn more about the Leo Rover's line follower example from this article.

Object detection

Leo Rover's object detection
Object detection from a Leo Rover’s camera

Another exciting feature we've prepared for your Leo Rover is the object detection example

It’s a technology related to computer vision and image processing that has to do with identifying and locating various objects, such as people, buildings, or chairs, to name a few, by displaying bounding boxes around the detected items allowing to locate their position. 

To implement the object detection example on the Leo Rover, we used a pre-trained model from the TensorFlow Lite repository. This model enables your robot to recognize a wide range of objects listed in the COCO dataset by drawing a frame around the detected item and displaying its label along with a confidence parameter specified as a percentage. Depending on your preferences, the parameter values can be easily altered by manipulating the sliders so that the feature draws bounding boxes around only those objects whose confidence meets the percentage threshold you've set.

Having your Leo Rover, a computer running on Linux OS or ROS, as well as some basic ROS understanding is all you need to fire this feature up. See how to do that in our tutorial.

ARTag follower

Leo Rover's ARTag follower example
A Leo Rover detecting an ARTag

And last but not least – the ARTag follower example. 

This feature will enable your Leo Rover to detect and follow a printed ARTag. The only things you’ll need are a Leo Rover (obviously), a printed ARTag, as well as the integrated software. 

Our ARTag follower example is based on a pre-made package from ROS for ARTag detection. It has specific parameters –the dimensions of 16x16 cm and ID 0. Only an ARTag with these parameters will be detected by this default feature. Of course, these parameters can be changed provided that you dig into and tinker with the appropriate files, and modify these parameters as you see fit.

To be able to run the ARTag follower feature properly on your Leo Rover, go to this integration.

Dive into Leo Rover integrations

The Leo Examples package will soon be a default part of LeoOS about which you can read here.

There are a lot of things you can do with your Leo Rover. Why don’t you check out other integrations available on our website? And we’re constantly working on new ones, so stay tuned :)

Want to see more posts like that?

Subscribe to stay informed.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

See more blog posts:

<- get back to the Blog