The best part: the training goes on even when you are away!

Researchers at Colorado State University have designed an artificial intelligence (AI) system that can help train dogs when their owners are away. Such a situation may well come around soon as people return to offices after vaccination against the coronavirus and life returns to normal. (Digital Trends)

The Colorado research team’s prototype device can shout out commands to your pooch, monitor if it is obeying them, and give a reward in the form of a treat if it does.

How AI-powered dog training works

“We have developed an apparatus which uses machine learning to monitor and reward dogs’ positive behaviors,” Tom Cavey, one of the co-authors on the project, told Digital Trends. “If a dog displays a desired action for an ample amount of time, the device will release a treat.”

At the heart of the device is a tiny, low powered, embedded computer called the Nvidia Jetson Nano.

The computer can monitor a live video feed of the dog’s doings, and with the help of machine learning, determine whether it has obeyed commands.

If the dog “sits” or “lies down” as instructed, the system detects good behavior and automatically drops a treat.

Note that the computer is completely independent and does not depend on any local or remote connectivity. Because of this constraint, the Colorado research team did not apply complex machine learning models that required more powerful hardware and memory. What they lost due to these limitations, they made up through the use of other strategies such as optimization and quantization techniques.

For overworked pet-owners, however, it reduces dog training to just loading up the treats in the device.

When can you buy the device?

The system is still in the prototype stage, but commercialization is a distinct possibility.

“Our approach consists of low-cost hardware and software components that would allow for the commercialization of a relatively affordable product,” said co-author Jason Stock. “We consider the possibility of developing this further given a greater budget for creating a better-functioning prototype, as well as improving the model and dataset to identify more behaviors—perhaps even those perceived to be negative.”

Related Story:  Boston Dynamics’ ‘Spot’ Robot To Work On Construction Sites